(navigation image)
Home American Libraries | Canadian Libraries | Universal Library | Community Texts | Project Gutenberg | Children's Library | Biodiversity Heritage Library | Additional Collections
Search: Advanced Search
Anonymous User (login or join us)
Upload
See other formats

Full text of "A history of oak furniture"

RO0€RT>)OlMeS 



FOR MAWV YEARS 4TEACHER 
IW THIS COLLEGE i«^75^K.£« 

THIS BGPK IS ONE 9FA Nil MB^'K 
FROM -jyiE LIBRARY p/Jvl'^ HOLMES 
PKESEMTED TOTHEQKITARIO 
COliEGE OF API DY HIS RELATIVES 
1U N E ^^^^^^^^193 O 



THE CONNOISSEUR SERIES 
OF BOOKS FOR COLLECTORS 

A HISTORY OF 
OAK FURNITURE 



I'l.ATE I. 




COURT CUPBOARD. DATED 1610. 

Victoria iind Albert Miiseiiin. 



THE CONNOISSEUR SERIES OF BOOKS FOR COLLECTORS 

EDITED BY C. REGINALD GRUNDY 



A HISTORY 

OF 

OAK FURNITURE 



BY 



FRED ROE, R.I., A.R.B.C. 

(Author of "Ancient Coffers and Cupboards," 
"Old Oak Furniture," "The Art of the 
Cofferer," etc.; Joint Author of "Vanishing 

England") 

Illustrated with Drawings by the 
Author, and from Photographs 



published bv 
THE CONNOISSEUR, 1, DUKE STREET, ST. JAMES'S, LONDON, S.W.I 

MCMXX 



lOHl 



J~3 



PRINTED BY 

BBMROSE & SONS LTD 

LONDON AND DERBY 




CONTENTS 



PAGE 

Introduction ._._----- 1 

Chapter I. — English Furniture from the Thirteenth to the End 
OF the Seventeenth Century .... - 4 

Chapter II. — English or Foreign? ----- 12 

Chapter III. — French Furniture to the End of the Sixteenth 
Century - - - - - - - - -15 

Chapter IV. — Furniture of the Netherlands - - - 19 

Chapter V. — Tyrolese Furniture ----- 23 

Chapter VI. — Some Fresh Figured Examples - - - 25 

Chapter VII. — Secret Receptacles- and Treasure Chests - 28 

Chapter VIII. — Furniture bearing Dates - - - - 34 

Chapter IX. — Personal Experiences - - - - - 41 

List of Plates --------- vi 



LIST OF PLATES 



PLATE 

Court Cupboard, dated 1610. Victoria and Albert Museum (/rontispiecf) .... I. 

Cupboard, temp. Henry VII., from Bur.varton, Salop (facing page 1) 11. 

Coffer of N. Fares, temp. Henry VU. (facing page 5) III. 

Northern French Coflfer, Fifteenth Century (facing page 13) IV. 

Normandy Stool, Fifteenth Century (facing page 19) - - V. 

Renaissance French Coffer (facing page 23) \'I. 

Walnut-wood Cabinet, French (Lyons), Sixteenth Century (facing page 29) ■ ■ ■ ■ VII. 

Oak Carvings from the late Sir Edward Holdcn's Collection (facing page 41) ... VIII. 

Types of Construction. (1) - - - - ■ - ^^■ 

Types of Construction. (2) X. 

Late Twelfth-century Coffer, in Hindringham Church. Norfolk XI. 

Thirteenth-century Coffer, in Boshani Church, Sussex XI. 

Panel from a Thirteenth-centui y Coffret. Saffron Walden Museum XII. 

Foot of the Boshani Coffer - . - - XII. 

Late Fourteenth-centurj- Coffer, in York Cathedral XI 11. 

Coffer of the Perpendicular Period, formerly in Sedlescombe Church, Sussex - - XIII. 

Late Fourteenth or early Fifteenth-century Coffer, in Chichester Cathedral . - - - XIII. 

Fourteenth-centui-y Coffret, and Sixteenth-centur) Chest XIV. 

Fourteenth-century Coffer, in St. John's Hospital. Canterburj .------ Xl\'. 

The Cradle of Henry V. London Museum XIV. 

Fourteenth-century Coffer, in Brancepeth Church. Northumberland XV. 

Late Fourteenth-century Box, Westphalian XV. 

Fifteenth-centurj- Panel, from York Cathedral XVI. 

Early Sixteenth-century Side-table ... - - XVII. 

Sixteenth-century Draw-table and Bench, from Broadway, near Ihninster .... XVII. 

Fifteenth-century Chest, French - XVI II. 

Late Fifteenth-century Cupboard XVIII. 

Early Sixteenth. century Dole Cupboard, from Ivy Church XIX. 

Chest, first half of Sixteenth Century, Westphalian XIX. 

Late Fifteenth or early Sixteenth-centuiy Cabinet XX. 

Sixteenth-century Box, covered with " cuir-bouiUi." Lincoln's Inn XX. 

Tyrolese and French Gothic Furniture - - XXI. 

Tyrolese Gothic Cupboards. Munich National Museum - - XXII. 

Late Gothic Coffers XXIII. 

Chest, temp. Henry Vlll. XXIV. 

Types of Tudor Flowering XXIV. 

"St. Augustine's" Chair, Canterbury - XXV. 

Chair-table, end of Fifteenth Century XXV. 



LIST OF ?hKTE%—contwued 



Gothic Joint Stool, in Brede Church, Sussex XXVI. 

Fifteenth-century Coffret XXVI. 

Joint Stool. Saffron Walden Aluseum XXVII. 

Front Panel of the Corporation Coffer, Ipswich XXVII. 

Gaming Table, 1530 XXVIII. 

Gothic Joint Stools XXVIII. 

"Grace Before Meat," by Jan Steen, showing a Linen-panelled Cupboard-seat - - - XXIX. 

Chest Front, middle of Sixteenth Century XXX. 

Linen-panelled Chest, French or Flemish, early Sixteenth Century XXX. 

Fifteenth and Sixteenth-century Panels, French - - XXXI. 

Middle Sixteenth-centurj- Panel, Northern French or Flemish ...... XXXII. 

French Credence of Renaissance Design XXXIII. 

Prior Silksted's Coffer, 1519. Shanklin Church XXXIV. 

Jongleurs car\ed on the Abington Abbey Frieze ......... XXXIV. 

Tudor Court Cupboard XXXV. 

Types of Tudor Panels XXXVl. 

Chest Front, French, temp. Vram^ois I" XXXVII. 

Cupboard, illustrating the Decadence of the Linen-panel XXX\'II. 

Italian Cassones - - - XXXVl II. 

State Chairs, French, Sixteenth Century XXXIX. 

Sixteenth-century Buffet, at Fontainebleau XL. 

Sixteenth-century Coffers, at Ingatestone and Framlingham - - XLI. 

Sixteenth-century Coffers, at Worcester and Winchester XLII. 

Alms-box, dated 1589. in Dovercourt Church, Essex XLIII. 

Secretaire, dated 1594 XLIV 

Details from Tudor and Seventeenth-century Furniture ........ XLV. 

Joint Stools XLVI. 

The Great Hall, Penshurst, showing Tables XLVII. 

Elizabethan Tables XLVIII. 

Elizabethan and Dutch Seventeenth-century Draw-tables XLIX. 

Apethorpe Hall, N'orthants, showing Tables and " 1364 " date on Screen . - - . L. 

Elizabethan Panelling of the "Nelson Room," Star Hotel, Great Yarmouth - LI. 

Inscription on the "Instone" Table at Denhani, Bucks LI. 

Details of Elizabethan Bedstead ...,- lU. 

Bedstead-head, first half of Seventeenth Century LIII. 

Elizabethan Inlaid Chest LIII. 

Sixteenth-century Table, at Sutton Place, Surrey LIV. 

Seventeenth-century Travelling Table LIV. 

Seventeenth-century Fireplace, at Shrewsbury LV. 

Key-hole .Mechanism of Dutch Cabinet LV. 

Seventeenth- century Cabinet, Dutch LVI. 

Elizabethan and Carolean Chairs LVII. 

Dug-out Coffer, in Bishop's Cleeve Church, Gloucester LVII. 

Seventeenth-century Bible-boxes - LVIII. 

Chair, dated 1649 LIX. 

vii 



LIST OF PLATES— fo«//wW 



PLATE 

Seventeenth-century Bible-box LIX. 

Chests, late Sixteenth and early Seventeenth Century LX. 

Flemish Cupboard, Seventeenth Century LXl. 

Spoon Cupboard and Knife-box - - LXll. 

Late Sixteenth and Seventeenth-century Chests LXIII. 

Seventeenth-century Marqueterie Chest - LXIV. 

Early Sixteenth-century Red Cedar-wood Coffer, Italian LXIV. 

Seventeenth-century Settle . - - LXV. 

Cradle, dated 1641 - - - LXV. 

The Fingringhoe Coffer, dated 1684 LXVI. 

The James Griffin Coffer, dated 1639 LXVI. 

Late Sixteenth or early Seventeenth-century Coffer LXVI I. 

Tyrolese Chest, dated 1624 - - - LXVIl. 

Chest of Drawers, second half of Seventeenth Century LXVIII. 

Spice Cupboard, second half of Seventeenth Centurj LXIX. 

Cabinet, inlaid with Mother-o'-pearl, etc., second half of Seventeenth Century LXIX. 

Small Square Jacobean Stool LXX. 

Inscription carved on Lid of the Rye Coffer, 1661 LXX. 

Seventeenth-century Chairs and Stool ... - LXXl. 

Seventeenth-century Chairs ... - LXXII. 

Day-bed, temp. Charles II. LXXIII. 

Late Seventeenth-century Chairs LXXIV. 

Joint Stool, dated 1689, Midhurst Church, Sussex LXXV. 

Lock-plate on the Ingatestone Coffer LXXV. 

Seventeenth-century Box, Dutch LXXVI. 

Fifteenth-century Carved Window-sill at Stamford LXXVI. 



NOTE. — All furniture figured amongst the illustrations is English unless otherwise described. A 
few articles in other woods than oak arc included on account of the very typical characteristics of 
their style. 

For a photograph of one of the late Sixteentli or early Seventeenth-century Chests, we are 
indebted to the owner, Mr. Thos. Scales Carter, of Oak House, llklcy. 



Plate II. 




CUPBOARD. TE.MP. HENRY VII. 

Victoria and Albert Musctnn. 



FROM BURWARTON, SALOP. 



INTRODUCTION 




j'HE elder Mr. Weller once observed that more widows were married than 
single women, and approximating with this cryptic utterance is the 
remark which I recently heard, " There's more old furniture existing 
nowadays than was ever made years ago." When properly unravelled, 
the latter assertion may be accepted as a truism. Other equally certain but less 
known facts also exist. Since the increase of publication of well-illustrated 
books depicting specimens of ancient furniture of unusual interest as regards form 
or structure, the reduplication of such types with a simulated appearance of age 
has become an active industry. It is seldom, however, that the copymg of such pieces 
from a pictorial illustration is anything of a real success, and if once placed beside 
the genuine article the difference in contour and quality would be both obvious and 
surprising. 

Occasionally it has happened that spurious (or mainly spurious) rarities have 
escaped challenge, and have appeared somehow in illustrated works on the subject 
of ancient furniture as articles of rare interest. These also, in spite of their ambiguous 
origin, have not escaped duplication, and the result has been a continuous declension, 
tending much to mystify the embryo collector with a little knowledge as to styles. 

Avowed and open reproduction of furniture on antique lines, manufactured from 
old beams and timbers reaved from barns and housebreakers' debris, is at the present 
time carried on with a good deal of ingenuity, and it is not infrequently remarked 
among the Philistines that such skilful imitations must detract from the value of 
original pieces. A greater fallacy never existed. Each one of these " faked " 
reproductions could be successfully refuted by the expert connoisseur, and as a 
parallel it may be remarked that the modern reproduction of ushabtiu and scarabs 
by thousands has never deteriorated the value of one single true piece dear to 
Egyptologists. 

The higher flights of forgery in all their various departments have for the last 
decade or so reached a pitch which it is difficult to define as either fine art or science. 
To accomplish that which shall successfully deceive the penetration of the cautious 
expert is a task which requires something more than ingenuity nowadays, and the 
superficial methods which imposed on human credulity some few years ago are being 
superseded in favour of systematical approach and a general thoroughness of execu- 
tion, the result of which it is sometimes exceedingly hard to unmask. But the 
difficulty of accomplishing^' successful detection in a really tough case is infinitely 
better for sharpening the wits of the collector than the steady and easy acquisition 
of true specimens of any class of antiquities whatsoever, and the resultant advantages 
are both lasting and sufficiently obvious. 

Forgeries may be roughly separated into two classes: (i) the forgery rank and 



2 A HISTORY OF OAK FURNITURE 

complete, and (2) the forgery which is a make-up or a composition of old and new 
artfully combined. Of these two species the latter is by far the most dangerous, 
being frequently a pretty object-lesson of the "half a lie" which is so hard a matter 
to fight. On the whole the French forger is more clever at the game than either his 
English or Belgian prototype. It is a well-accredited fact that the best of the " fakers" 
in France take the trouble to reproduce the tools of some three or four hundred 
years ago before proceeding to attack the actual wood in their productions. Wear 
is not simulated by the use of sand-paper — a very inadequate means, but created 
by the constant rubbing of greasy cloths on such portions as would be most likely 
to be affected. I have known of an old soldier, badly in want of work, who was 
employed by an unscrupulous firm to dust some white powder over his clothes, and, 
after having seated himself in the reproduction of an elbow chair, and finding out 
from the marks where most wear was likely to come, to have applied himself to the 
impairing of such parts with coarse sebaceous plush torn from old carriage linings. 
But forgeries in a certain respect resemble the deeds of other criminals; the small 
but infallible incongruity or omission which gives away the most artistic crime is 
seldom absent, and, once descried, can never be misunderstood. One of the very 
best counterfeits in old oak, as regards surface, which I have ever seen, was primarily 
detected by the carver having mixed up details of the thirteenth and late fifteenth 
centuries (each admirable in their way) in a most impossible manner. It is best not 
to particularise; this wonderful production may at the present moment be the pet 
object of some so-called antiquary's collection, and to disillusion the owner would be 
cruel, besides possibly leading the way to complications. 

When discriminating about old furniture, one must never forget that farmhouse 
pieces were always intended as such ; that is to say, they were heavy, substantial, 
perhaps boldly or roughly-carved specimens made for the yeoman or middle class 
for which they were origmally intended. The well-bred gallants and doubtful 
ladies of Charles's court, who masqueraded as swains and shepherdesses, were no more 
the real article than the rugged, carved armchair was the type or mould which was 
fashionable in more elevated circles. Marie Antoinette playing at haymaking at 
the Petit Trianon is not to be compared to the woollen-yarned and buskinned bucolics 
whose veritable labours were mimicked so lightly. And yet, both as a paradox and 
a parallel, how many descendants of ancient families can be traced into the humblest 
of surroundings, even as court furniture, through nameless vicissitudes, has often 
found its abode in the dwelling of both the yeoman and the tenant farmer side by 
side with the substantial and more coarsely-carved pieces made for that class of the 
community. But it is not difficult for the connoisseur in process of classification to 
divide the higher patterns from their more humble prototypes, always remembering 
to make allowance for an advance in ^he change of style in the case of court pieces, 
sometimes as much as over forty years' difference being observable between objects 
of similar use that were actually made at one and the same time. Other instances 
are even more discrepant, and a late country-made piece may at first sight exhibit 
superficially earlier characteristics than its town-made brethren, which can be disproved 



A HISTORY OF OAK FURNITURE 3 

on close and thorough examination. Such examples are not unfrequently met with. 
Let us take a few examples and compare them — 

(i) A chest of severely architectural design bearing a date about the middle of 
the sixteenth century. The piece is modelled on classic lines, and unques- 
tionably belongs to the year inscribed on it, but a casual observer would 
assign it to a date not earlier than the early part of the seventeenth century. 
An exceedingly courtly and well-designed piece in advance of its time. 

(2) A chest from the West Country, of an earlier style than the last, carved with 
medallions and heads in profile, and, roughly speaking, representative of 
a certain style that prevailed during the reign of Henry VIII. A date 
(during the first quarter of the seventeenth century) which is incised upon 
the top transom has been challenged as an addition made by an owner 
subsequent to the time when the chest was made, but careful examination 
discovers a thinness of the uprights and certain peculiarities in costume 
which undoubtedly place this particular article as belonging to the date 
which appears on it. Ergo, a countryside piece exhibiting lingering tradi- 
tions, and behind the times as regards style. 

Such curious discrepancies are at first sight rather puzzling, but they belong to 
a distinctly different class to the anachronisms perpetrated by many modern carvers, 
and their discovery leaves no shadow of doubt on the veritable piece which exhibits 
them. 

A few of the pitfalls for the unwary have been touched upon here, but for 
those who run and read as well, the acquirement of antique oak is a pursuit which 
increases in ardent interest as the quarry grows scarcer. 

FRED ROE. 
18, Stanford Road, 

Kensington Court, W. 8. 





CHAPTER I. 

English Furniture. 

From the Thirteenth to the End of the Seventeenth Century. 

[HE history of furniture is in many respects a record of the architectural 
styles which have prevailed in the country to which such furniture 
belongs, and without some auxiliary knowledge of architecture it is 
impossible to develop any sound attainments on the subject of antique 
furniture. Again, some knowledge of costume is occasionally very helpful in 
determining the ages of debatable pieces bearing representations of the human 
figure, though discoveries of these rarities occur only too seldom. As regards early 
furniture made from wood, it appears as though oak was the sole material used, 
the softer trees being ignored, or else we must conclude that any articles made from 
them in Gothic times must have perished. The earliest pieces of furniture con- 
structed were doubtless chairs and coffers, or boxes of some sort, and as these must 
have been originally intended for hard wear and strength, the tougher material was 
probably invariably used by choice. As instances of what solid construction com- 
bined with heart of oak may effect, a great number of thirteenth-century coffers 
remain in this country to testify. These receptacles, which exist mainly in churches, 
were so hardy and indestructible that with the slow increase of the population it 
was apparently unnecessary to construct but few pieces of a similar kind in the two 
succeeding centuries, the older coffers remaining in use long after the craftsman who 
made them had gone to dust. This is the most potent reason I can advance for 
the numerous examples of thirteenth-century boxes yet existing compared with the 
very scanty number of such pieces belonging to the two succeeding centuries. 

Eliminating archaic relics of pre-Norman origin, it is possible that the earliest 
coffer remaining in England which is decorated with any carving is that in Hindring- 
ham Church, Norfolk. An arcade of interlaced semi-circula ■ arches of the Transition 
Norman period is incised on the front panel and uprights of this probably unique 
piece, which can hardly have emanated from a later period than the last quarter of 
the twelfth century. A coffer in Graveney Church, Kent, which exhibits incised 
carving of a similarly early kind, follows close upon its heels; but as the arcading in 
the Graveney example is composed of pointed and cusped arches of the subsequent 
Early English style, it is obvious that the priority of some years must be given to 
the Hindringham relic. 

Coffers of this period, down to the end of the thirteenth century, were not 
provided in the ordinary way with hinges, the lids revolving on iron pins which were 
inserted horizontally through the back uprights, while the rear panels were often 
protected by chains. It has been conjectured by some, from observing fragments of 
these chains remaining, that they were attached for the purpose of fastening such 



,.*"•' 



Plate III. 




z 

a 
I 

a 
s 

a 



en 

a: 



Li. 
U. 
O 



A HISTORY OF OAK FURNITURE 5 

coffers to the wall, but this is a fallacy, as their intention was really the safeguarding 
of the receptacle from illicit attack by strengthening the connection of the lid with 
the body. The uprights of these coffers are invariably of great width, and are 
frequently termed standards. A thirteenth-century coffer of small size exists at 
Pitstone, Bucks., in which the standards are wider than the central panel. 

Thirteenth-century coffers which exhibit elaborate carving or geometrical roundels 
incised on their fronts exist, among other places, at the churches of Saltwood, Kent; 
Stoke d'Abernon, Surrey, Earl Stonham, Suffolk; Westminster Abbey, and York 
Cathedral. Sussex is also particularly rich in thirteenth-century carved coffers, notably 
those remaining at the churches of Buxted, Clymping, Felpham, Alidhurst, South 
Bersted, and Chichester Cathedral. 

Coffers bound with iron scrollwork, and approximating to the same period, may 
be found in a fair amount of English churches, those at Brampton, Northants; 
Chobham, Surrey ; Icklmgham, Suffolk ; Salton, Yorks. ; and Wootton-Wawen and 
Rugby, Warwickshire, being among the most notable. The last-mentioned example 
is mounted on wooden wheels for purposes of transport. 

English coffers belonging to the fourteenth century are much scarcer than the 
foregoing, but beautifully carved specimens remain at Brailes, Warwickshire; Der- 
singham and South Acre, Norfolk; All Saints' Church, Hereford; Litcham, Norfolk; 
and St. John's Hospital, Canterbury. Other fine patterns of this period may also be 
found, but in many cases they have been mutilated by wilful and ignorant persons. 
Lincolnshire, perhaps, possesses two of the most complete fourteenth-century relics of 
this class in the coffer at Haconby Church and the chest at Huttoft. I use the 
words " coffer " and " chest " in the strictly architectural term, as applied to boxes 
composed of single panels, as compared with those constructed with multiplied 
framing. Most of the examples mentioned exhibit the winding tracery peculiar 
to the Decorated style — a continuous flow of curvilinear lines which trail into each 
other without the stops and uprights inaugurated by the succeeding Perpendicular 
method. Chairs or benches of the thirteenth and fourteenth centuries, belonging to 
anything else than church furniture, hardly exist in England, and those which do 
remain can in nearly every instance be recognised as fragments of stalls; but a few 
standing cupboards of heavy build continue with us, their doors bearing strap hinges 
terminating in an ornamental finish. A fine specimen of this description is preserved 
in Chester Cathedral, and may be accepted as the type of cupboard generally in 
use among the better classes during the thirteenth and fourteenth centuries. 

English coffers of the fifteenth century are even rarer than those of the preceding 
epoch. A few iron-bound receptacles belonging to the period exist, but those bearing 
any representation of the tracery of the period are scant enough. The churches of 
Barrow-on-Soar, Leicestershire; St. Michael's, Coventry; and perhaps Little Walding- 
field, Suffolk, furnish the most notable examples. Flemish chests, such as those of 
St. Mary's Church, Cambridge, or Southwold, Suffolk, may date from this period, 
but they do not represent the national art and craft which we are considering. Some 
hutches and cupboards pierced with late Perpendicular tracery have been discovered 



6 A HISTORY OF OAK FURNITURE 

during recent years, but these are food receptacles, mostly of rough make, and from 
their decoration belong to the borderland of the fifteenth and the succeeding century. 

During the reign of Henry VII. furniture began to make a distmct advance, 
both as regards design and construction. That we were later in our developments 
than most European countries there is no doubt whatever, but it is also quite certain 
that we preserved a good many of our national characteristics, even throughout the 
changes effected by the Renaissance. Material was lavishly employed in our home- 
made productions, and the bold cutting of the English craftsman was distinct from 
the florid heaviness of the German, or the finely-traceried productions of the French. 
Flemish work affected us, but during the declining Gothic style its bias was rarely 
felt inside the East Coast, and was evanescent. In the style which is coeval with 
the first half of the sixteenth century, the Flemings asserted themselves more fully, 
and were much patronised in Court circles and by the wealthy for their products; 
but their mode was not really an original one, consisting as it did of the borrowed 
revival of Italy affected by their own somewhat ponderous temperament. Abun- 
dant proof that the Flemings were more successful here in their quasi-Roman pro- 
ductions than in their earlier style may be found in the evidences of their hands, 
which exist all over the country in the shape of screens and fixed woodwork generally. 

We get the earliest distinct evidences of English domestic furniture in pieces 
belonging to the reign of Henry VII., which have existed down to the present time, 
and on these we may find Gothic tracery of a debased type, insidious suggestions 
of Italian detail, and the linen panel.* Such decorations, however, run more abun- 
dantly through the succeeding reign, and it must not be assumed, if any of these 
characteristics appear in conjunction with each other, that the piece bearing them 
must necessarily belong to the first Tudor's reign. Only discrimination will determine 
that, and even then but rarely. It is likely that the greater portion of the furniture 
made for domestic purposes was cumbrous and unadorned with carving, and that it 
was broken up for the sake of the material when newer modes came in. Feudal 
customs had not yet departed, and huge trestle-tables with independent tops were 
still in use in the halls of the great, while forms and benches constituted the major 
part of the seating accommodation, only those personages on the dais being provided 
with chairs. 

Side-tables and credences were more elaborate, and the best pieces presented an 
effect in which exquisite design is tempered with restraint. Such examples are 
extremely rare, but our Victoria and Albert Museum possesses a specimen of English 
art of the end of the fifteenth century which it would be difficult to excel. t This is 
an oak side-table of somewhat low dimensions, having a shallow cupboard beneath, 
and its front panels perforated with devices, among which appear the sacred 
initials " I.H.S." surmounted with a crown, the arms of FRANCE MODERN, the Tudor 



* Portrait heads in profile were not enclosed in roundels till the reign of Henry VIII., when the 
Renaissance was established, 
t No. W. 47—1910. 



A HISTORY OF OAK FURNITURE 7 

rose, and a window of late Gothic tracery. In the sides are linen panels placed 
longitudinally, and the space for a door, which is now missing, appears in the 
centre of the cupboard beneath the table. This beautiful and interesting piece was 
evidently made for some personage of high rank, and was probably used for the 
carving of food and its temporary holding. 

A few chairs and stools of the same period are left to us. One very perfect 
linen-panelled specimen is at St. Donat's Castle, in the collection formed by ihe late 
Mr. Morgan Williams, while an excellent range of slab-ended stools of the keyed 
type has been gathered together within recent years by the Marquess of Granby. Of 
the subsequently developed pattern of " jo\'ned," or joint stools. Sir George 
Donaldson possesses some remarkable examples. 

In 1509 great interest was aroused by an announcement which was made in the 
daily papers that His Majesty (the late King Edward VII.) would use in one of 
his levees an historic oak chair, an heirloom of a Warwickshire family, and which 
was traditionally supposed to have been used by Henry Tudor, Earl of Richmond, 
afterwards King Henry VII., at a council of war held on August 20th, 1485, at the 
Three Tuns Inn, Atherstone, where he spent the night. How the chair came into 
the possession of the family is not quite certain, but it is supposed to have been 
purchased some three generations ago, when the said inn was pulled down. That 
the chair was an old one there was no doubt about, but on the publication of a 
photograph it at once became apparent that the chair for which so interesting a 
history was claimed could not possibly be older than the seventeenth century. The 
back was panelled and carved with an arch enclosing a diamond, in true Jacobean 
style, and the turned legs were utterly unlike anything imagined or executed by 
craftsmen of the time of Henry VII. This incident will be fairly fresh in the 
minds of students of early furniture. 

A supposition attached to an oak cupboard in the Victoria and Albert Museum 
has more possibility about it. This piece, which is a tall standing press of rough 
construction, is fashioned in the flat shape associated with the reigns of the two 
last Henrys. It is, by the way, wrongly described as a " Livery Cupboard," and 
is supposed by some to be connected with Prince Arthur, eldest son of Henry VII., 
who died in 1502, in his sixteenth year. The front of this cupboard exhibits a 
number of carved rosettes and wheels, and the doors to its two stages are placed in 
the centre, as was invariably the case about this period. The top door and its 
attendant side panels are pierced with windows exhibiting Gothic tracery of a late 
type, that in the door being pointed, and those in the panels being flat topped. 
There is absolutely nothing about the cupboard to suggest that it belongs to any 
other period than the epoch named, and it is equally certain that it is of English 
make; but the attribution of its ownership, which is based on certain decorations, 
hardly proves conclusive. Right and left of the door on the lower stage appear 
certain perforated decorations, which, it is asserted, represent the ostrich feathers 
appertaining to the Prince of Wales. The matter is quite a debatable one, and 
depends in great measure as to the real meaning of these devices. For my own part. 



8 A HISTORY OF OAK FURNITURE 

I am not sufficiently impressed with the hypotliesis of its princely attribution, which 
has the disadvantage of being open to some grave objections, too lengthy to be dis- 
cussed here. This press is an exceptionally interesting piece, and the accuracy or 
otherwise of the foregoing legend makes no difference to its antiquarian value; but it is 
doubtful whether such a roughly-made article would be built for so high a personage 
as Prince Arthur, and with all reservation I must state my belief that the devices 
represent in some way a rebus on the name of the original owner, or an intimation 
as to his calling or pursuits.* 

Another domestic piece of the same period in the Museum is an oak coffer, 
carved on its obverse with the conventional rose-bush and flowering, and on the 
reverse with a vine-trail and the inscription " N. FARES," accompanied by an object 
which looks suspiciously like a skull-cap. The reverse side is flanked with buttresses, 
and on one end alone appears the monogram " N.F.," surmounted by a similar skull- 
cap. The lid is very massive and weighty, and bears on its fore edge a beautiful 
Gothic moulding. To my mind this intensely attractive piece must have been the 
business coffer of some leech or apothecary of high standing. The end destitute of 
carving would be placed against the wall, and on the opening side of the coffer 
(which would also serve as a counter) Mr. Fares would sit to dispense his prescrip- 
tions, while on the other (that adorned with buttresses) his patients would be con- 
tinually reminded of his honoured name. The carved monogram on the side would 
be visible to those who passed the counter into the inner sanctum for a private 
consultation. This unique piece was formerly elevated on legs, and a canopy of 
depressed Gothic arches appears between the buttresses on the side which would 
face the public, as well as a single arch beneath the monogram on its carved ends. 
No arches were placed on the side which Mr. Fares would sit at, the omission being 
obviously for comfort in accommodating his knees beneath the body of the coffer. 
I know of no domestic piece which possesses a deeper interest in suggesting a history 
of its primary uses. 

It is curious that the early cabinet-makers should have had so little of the 
utilitarian about them that they persisted in placing the small doors of their cup- 
boards invariably in the centre of the piece, an inconvenient arrangement, which allowed 
the users to grope about blindly in corners, with but remote chance of finding what 
they required. English cupboards of late fifteenth and early sixteenth century 
design were without prominences and wanting in contour, for, while assimilating 
the features of the Renaissance, the bold ornate outlines of the Italian cassones 
did not seem to have been followed by our English craftsmen. Cornice mouldings 
were flat, stiles were wide, and panels often unnecessarily small. We occasionally 
find cupboards and chests of the time of Henry VIII., the plain contour of whose 
fronts is broken by heads in high relief, boldly projecting from the centre of the 
panelling, but it was not till the age of Elizabeth that any leaning was shown towards 

• No. W. 15 — 1912, Victoria and Albert Museum. Presented by Robert Mond, Esq., F.S.A., 
through the National Art Collections Fund. 



A HISTORY OF OAK FURNITURE 9 

canopied recesses and baluster supports. The forms of decoration employed in 
England during the first half of the sixteenth century were arabesques, pilasters, 
and grotesque heads enclosed in roundels. Henry VIII. greatly encouraged foreign 
craftsmen, and there is no doubt that much of the woodwork executed here in his 
time was carved by Flemings, Germans, and Italians. A few of the English crafts- 
men persisted m their endeavours to perpetuate the moribund Gothic style, such 
efforts being, properly speaking, a species of dogged defence or inability to assimilate 
more fashionable modes, which was peculiar to our own country alone. Some curious 
instances of such contrarieties are noted in the chapter on " Furniture Bearing Dates." 

Large tables and forms were seldom decorated with carving in our own country 
about the time of Henry VIII., such aesthetic qualities as they possessed being afforded 
by the outline alone. The quaint early custom of having connecting tie-bars running 
underneath the centre of the table instead of foot-rails, and keyed at each end by 
means of wedges, was adhered to in great measure, and most of these tables, which 
are classed as " Gothic," belong really to a period subsequent to the age which they 
are supposed to represent. 

Small occasional tables for the purpose of playing at cards and other games 
were, unlike their larger brethren, highly enriched with carving. Shaw gives a very 
careful engraving of one of these in his book on " Ancient Furniture," and mentions 
it as being in the possession of Mr. Swaby " from Hill Hall, Essex."* A table 
similar to the foregoing was exhibited at the Franco-British Exhibition in 1908, and 
was so remarkably like the specimen in Shaw's book that I am tempted to think it 
may be one and the same, though without having had an opportunity of closely 
examining it. If not the identical specimen, it at least must be from the hand of 
the same craftsman. A chair of precisely the same period, and decorated with 
roundels and ornaments of exactly the same character, is also figured in Shaw's 
work, the following description being attached to it: — "Chair of the time of Henry 
the Eighth, in the possession of Joseph Abel, Esq., Surgeon, Mitchel Dean, Glouces- 
tershire. This interesting example is of a period after the introduction of what was 
termed ' Romayne work ' into England ; that is, when the frescoes adopted by Raphael 
and other celebrated painters from the antique had given a new character to orna- 
mental furniture. It is so carved within, and with napkin panelling without, is of 
oak, and in a good state of preservation." 

Writing-desks with slanting lids do not appear to have come into use till after 
the mid-century was passed ; that is, desks which were constructed so as to be 
movable and not fixed like lecterns. The earliest decorated pattern I can point to 
in one of these movable desks is in the possession of T. C. Parker, Esq., of Skirwith 
Abbey, Cumberland. It was picked up in a Westmoreland village, and was supposed 
to have been originally in the parish church. This piece, which is carved with vine 
tendrils and thistles, also exhibits profile masks on its sides, the vine-scrolls issuing 



* specimens oj Ancient Furniture, by Henry Shaw, F.S.A. 1836. See plate 28. 



lo ^ A HISTORY OF OAK FURNITURE 

from the mouths of the masks. The piece is decidedly unusual, and I should hesitate 
to place a later date on it than the first ten years of Elizabeth's reign. 

It is difficult to assign any furniture particularly to the reigns of Edward VI. 
and Mary ; their respective reigns — six and five years- — were too short to admit of 
any specialisation. The wooden arcaded superstructure on Edward the Confessor's 
shrine in Westminster Abbey, attributed to the reign of Mary, is a somewhat severe 
piece of classic, which differs essentially from the grotesque vagaries produced in 
her successors' days. I have seen at least one piece of furniture bearing an inscription 
attributing its origin to the reign of the last Tudor king, but I should be somewhat 
chary of pronouncing upon the genuineness of either inscription or the article itself. 

The art of inlaying with different-coloured woods became fashionable during 
the sixteenth century. This form of embellishment emanated from Italy, and, spread- 
ing through France and Germany, speedily became very much in favour. It is not 
unusual to find the inlaid decoration on German chests so closely resemble that on 
our home-made pieces that it is difficult to believe the work could not be executed 
by the very same craftsmen. In most cases, however, the German article will be 
found to antedate the English, however closely their respective inlay may approxi- 
mate. During the latter half of the sixteenth century conventional floral decoration 
was much in vogue with the inlayers, but the simulation of quartered panels and 
geometrical insertions was also carried out. Painting and gilding was also resorted 
to, though not so freely as in the Gothic times. An item in the Kenilworth Inventory, 
taken in 1584, mentions "a bedstead of walnut tree, toppe fashion, the pillars redd 
and varnished " ; and a few articles are still occasionally found which exhibit traces 
of their embellishments in colour, though this has mostly disappeared under the 
efforts of the owners to promote a rich polished surface. 

Carved terminal figures were first introduced into English furniture during the 
third quarter of the sixteenth century, on the stiles of chests, cupboards, and the backs 
of bedsteads. These figures, far from being of a classic nature, as in Italian 
specimens, were frequently uncouth, and represented as wearing barbaric head- 
dresses. Cupboards assumed greater variety of contour, and were in part recessed, 
or had their sides facetted in a manner peculiar to productions of the English nation. 

Tables of Elizabeth's reign were made in a variety of forms, occasionally being 
of an architectural character, but more frequently having the legs formed in the 
shape of huge bulbs decorated with jewel moulding — a complete change to any 
preceding style. 

> With the advent of the Stuart dynasty, the history of furniture becomes less 

complicated. Classic influences gradually clarified during the later Renaissance, 
and though many of the mannerisms of Elizabeth's days were primarily adopted, 
the Jacobean style, on the whole, progressed in purity. One of the minor exceptions 
to this as regards surface decoration was in the form of ornament known as strap- 
work. Jacobean strap was almost invariably of a more debased shape than that of 
Elizabeth's time, the terminations often splaying into more or less meaningless forms. 
Whilst the craze for collecting rare bulbs prevailed in Holland, it was a favourite 



A HISTORY OF OAK FURNITURE n 

practice to introduce the tulip into decorative carving. This is an important point 
to consider since English furniture displaying the tulip can nearly always be assigned 
to no earlier period than the middle of the seventeenth century. Such articles as 
chairs, cupboards, and chests remained substantially the same in outline till the 
Dutch influence introduced in the reign of Charles II. brought a new mode into 
Court circles. Chests of drawers became more fashionable than chests, being 
frequently decorated with raised and facetted panels in the Flemish style. With 
the increase of printing, Bibles became more plentiful, every homestead of any 
standing possessing a family Bible with its accompanying Bible-box — in reality only 
a late survival of the coffret. And as feudal customs declined the massive tables 
which consumed so much of England's ancient oak became replaced by that national 
product now known as the " gate table." As the Jacobean period progressed, taste 
verged towards lighter construction. Soft woods in a great measure replaced oak 
as a material, preparing the way for mahogany in later years. Oak furniture, how- 
ever, continued to be made in outlying districts well into the eighteenth century, 
though merely as a decadence of previous types. The one attempt at a definite 
resurrection of accepted oak styles occurred in tlie time of Sir Walter Scott, and the 
result was not altogether a happy one. It is a singular fact that undoubted speci- 
mens of "Abbotsford " furniture may be seen at the present day in at least one great 
public collection, being described as dating from the second half of the seventeenth 
century. It requires little expert training to distinguish these meaningless and heavy 
attempts at reproduction from the true articles which were produced at the time of 
Britain's merriest monarch. 





CHAPTER II. 

English or Foreign ? 

'HERE is a distinct disposition among certain present-day collectors and 
critics to class almost all specimens in England antedating from our 
Elizabetiian style as being of foreign origin, and about this impression 
there is a good deal to be said. In the first place it is well to enquire 
what is actually meant by the word " foreign," this appellation being frequently 
used in such an indiscriminate way that some investigation should fairly be 
made as to the sense in which it is employed. Does it signify that the piece so 
referred to was designed abroad, or made abroad, or both designed and made in 
some country other than England ? I have heard dozens of critics who, because 
they have noticed certain lines and peculiarities in early coffers which exist or have 
been brought from the Continent, have not hesitated to sweepingly assign an origin 
French, German, or otherwise to undoubtedly English examples remaining in sitit 
in our own country. It is well known that in the thirteenth and early fourteenth 
centuries certain common characteristics in architecture prevailed more or less over 
a great part of Europe, and often it is not only difficult, but more than difficult, to 
" place " an early piece from its actual lines or decorative carving alone. The 
armoire at Aubazine, in Correze, is without doubt a French example, but there is 
nothing to distinguish it from Anglo-Norman design of the same time, if indeed 
we possess any so perfect examples of a like date. Again, the coffers at Brancepeth 
and Wath, indubitably English examples, had rightly never been queried till the 
recent importation of a German coffer designed on somewhat similar lines raised 
the hypothetical theory as to their possible origin. The similarity decided the 
matter in certain quarters, and it was urged by some that the English examples 
necessarily emanated from a German source. This, however, is hardly conclusive. 
There are such questions as the nationality of the wood which the pieces are con- 
structed from (German oak differs materially from English), the handling, construc- 
tion, and .several other points which apparently have not been considered. Lastly, 
but not least, there remains the fact that designs were more freely interchanged 
during the Middle Ages than is generally imagined, and though sometimes they may 
not have been adopted in the fullest sense by native families practising their crafts, 
these designs may have influenced some who would not care to adopt them in their 
entirety. The English craftsman was often bold and coarse, the German frequently 
heavy, and it is sometimes a delicate matter to decide authoritatively upon the origin 
of certain pieces without the most scrupulous examination. The coffer recently 
acquired by the authorities of the Victoria and Albert Museum (No. V., 49, 191 2) 
bears an extraordinary superficial resemblance to its prototype in Saltwood Church, 

12 



Plate IV. 




M 
'J 

I 
H 
Z 

£- 



CL, 

; 

to 

^ I 

Z "S 

^ I 

1 -2 

O i! 

z :^ 



A HISTORY OF OAK FURNITURE 13 

Kent; but close scrutiny will discover a certain lumpiness and heaviness of execu- 
tion very different from the Kentish example. It is well to note that the Kensington 
coffer came from Cologne, and is stated to have been found in a neighbouring farm- 
house on the Rhine, also that the material from which it is constructed is German oak. 

One thing must be admitted : it is more than singular and worthy of remark 
that such striking similarities could e.xist in work produced b)- different countries 
in a period so remote as the thirteenth and early fourteenth centuries, while a hundred 
years later, when communication must necessarily have developed, divergencies in 
style and character had matured to a remarkable degree. There were few European 
countries, with the exception of the Scandinavian group, that did not produce directly 
distinctive types of furniture m the fifteenth century. The variance had by this 
time become conspicuous. France and Flanders to a certain extent are exceptions 
to the rule. While it is easy to distinguish between French, German, and English 
examples of the fifteenth century, it is often a task of some difficulty to differentiate 
between French and Flemish examples, a slight heaviness in the latter being at times 
the only observable distinction. In later periods this heaviness became accentuated 
more fully, and frequently obtrudes itself in an unpleasantly redundant manner. 
Sir Walter Scott hits off a description of such production rather neatly in the follow- 
ing sentence, quoted from Paul's Letters to his Kinsfolk : " The same cumbrous 
solidity which distinguishes the female figures of Rubens, may be traced in the 
domestic implements and contrivances of his countrymen." 

Similarly, when one examines antique Flemish carvings of human figures, 
whether statuettes or in relief, we find that the countenances, especially of the women, 
are portrayed in a stolid, unintellectual cast. Virgin Marys, saints, and mythical 
characters alike sharing the prevailing ponderosity. The Flemings, as much as any 
race on earth, adhered to depicting their own immediate types and surroundings, 
and any lightness or grace which they possessed was most certainly derived from 
their French neighbours. 

One of the most instructive pieces which it would be possible to study in the 
debatable class is a coffer of carved elm in the Victoria and Albert Museum (W. 30, 
191 3). This coffer is sculptured on its front with a design of gryphons and scroll- 
work so unmistakably Tyrolean in character that few would hesitate to pronounce 
in this wise on its origin from superficial scrutiny. Running round the base, however, 
appears the confusing inscription : — 

BY . JAMES . GRIFFIN 
16 . THIS : CHEST . WAS : MAD : IN : THE : YEARE 

OF . OUR . 39 
LORD : GOD . ANO : DO. 

The initials " I.G." also appear in smaller characters in the centre of the front 
panel. Now here is a piece quite Tyrolean in character, and yet made by a craftsman 
who obviously worked in our own country, and, eschewing oak, worked in the softer 



14 A HISTORY OF OAK FURNITURE 

elm for his material. The peculiarities are too strong to admit of any doubt; the flat- 
surfaced carving is essentially Tyrolese in character, though perhaps the diamond- 
shaped ornaments scaircely accord; and the very red and black in depressed portions 
of the carvings is characteristic of the Tyrol. It is, therefore, extremely probable 
that James Griffin was an immigrant, long resident in England; that the coffer in 
question was made here for his own family use in the style which he remembered 
and loved; and also remotely possible that the gryphons (though the device was 
quite an ordinary one) were intended as a rebus on his name. The material used 
(elm) must have been selected as being more akin to the consistency of the Alpine 
fir than our iron-grained national oak. Altogether a very pretty piece of conjectured 
history. 

Another article of furniture in the same museum, but of a very much higher 
class than the last, is a cabinet of pinewood, elevated on pillared legs, the whole 
piece being of an architectural design. It is adorned with a number of boxwood 
carvings in high relief, representing battle scenes, while the flat surfaces are pro- 
fusely overlaid with designs in marquetry of sycamore, pearwood, Hungarian ash, 
maple, and other woods. On the base and bordering the inlay appear the rose and 
portcullis, emblems of the Tudor sovereigns of England. This elaborate piece of 
furniture, which dates from the middle of the sixteenth century, has nothing in 
common with English work of that period, and was most probably made by a North 
Italian craftsman, either for presentation to some personage of high rank in England, 
or else by his or her order. The front portion of the frieze is capable of being raised, 
disclosing five secret drawers, which, however, are carved in unison with the rest of 
the piece.* 

* No. 27, 1869, Victoria and Albert Museum. 





CHAPTER III. 

French Furniture. 

To THE End of the Sixteenth Century. 

DESCRIPTIVE syllabus of the various changes that took place in 
French furniture might well occupy more space than any chapter in 
this work, so much fancy and originality were exhibited in their various 
developments. The earlier styles, however, can only be dealt with in the 
present section, reserving the more recent evolutions for a future dissertation on the 
subject. As regards the earliest known styles of furniture, pretty much the same 
may be said of French examples as of their English contemporaries. Both con- 
struction and decoration are alike in each case, the singularities being oftentimes 
strikingly similar. After the thirteenth century, however, changes began to manifest 
themselves, and by the middle of the fifteenth century, when the Flamboyant style 
was fully established, the changes between our national productions and those of our 
French neighbours were very marked. We have little left of our own household 
movables belonging to this period which can compare with the French for delicacy 
of carving and precision in the matter of joinery. 

In France nowadays the pointed styles are more generally known and appreciated 
than in England. Not only the middle classes, but the common crowd there are 
conversant with the st)-le and term " La Gothique " to an extent unknown on this 
side of the Channel. Our own countrymen, with the exception of the architectural 
profession and a relatively inconsiderable number of artists and archaeologists, are 
by far the greater part either carelessly indifferent to or else ignorant of the 
phases of this most beautiful art, and an overwhelming majority even of its actual 
designation. It is remarkable, too, that among those here who are lovers of the 
pointed styles of architecture in stone buildings, a comparative few seems to apply 
the same interest to woodwork of these periods. In England Gothic art is for the 
few, at least as far as furniture is concerned, and all the transient and, in many 
cases, misdirected efforts of Walpole, Pugin, Shaw, and Sir Gilbert Scott have not 
succeeded in re-establishing any wide appreciation of it. 

It must not be supposed when one uses the denomination " French " that it is 
intended to include in one class all periodical similarities from the Pas de Calais 
to the Pyrenees, or from the Alps to the Cotes du Nord. Provincial differences were 
just as much varied across the Channel as the discrepancies between our craftsmen 
of Yorkshire and those of the East Coast, or even more so. Between Norman and 
Breton work a wide gulf is fixed, and during the latter half of the sixteenth century 
an even wider divergence existed between the productions of St. Lo and Lyons. 
This may be partly explained by the resistless push of Italian influence on an 
art-loving nation such as the French have always been. Italy was conquering with 

IS 



i6 A HISTORY OF OAK FURNITURE 

her arts more than with her armies. Yet the best workmanship must greatly depend 
upon the various styles which periodically reigned in separate provinces. Personally, 
I am of opinion that though the finest specimens of Gothic may have been produced 
about Rouen, a chaster and purer classic emanated from towns such as Lyons, Nismes, 
or other places in the Rhone Valley, where the old Roman influence was yet strong, 
and where the tide of Italian invasion was more speedily and strongly felt. 

At present our attention is on an earlier style — that Gothic which gave Rouen the 
famous rose-window in its west front, and the church of St. Ouen in all its complete- 
ness of style. French furniture of the fourteenth century is scarce enough, like its 
English prototypes, but of the following epoch there are plenty of examples. While 
English productions of the fifteenth century are infrequently met with, and in any 
condition may be regarded as rarities, there is no lack of French household movables 
of this epoch, they being mostly decorated with the graceful Flamboyant carvmg which 
attained to such a pitch in the fifteenth century. Our own local museums seldom 
enough make any show of antique furniture, but each little dusty public collection 
of curiosities in provincial France almost invariably contains some treasures in this 
way, and amongst them a Flamboyant chest is rarely wanting. Eliminating the 
suspicion that the disastrous effects of the British climate may in some part account 
for this discrepancy, there can be little doubt that at this period the French nation 
was more prolific and luxurious than the English. In spite of the destruction caused 
by time and vandal's hand, modern France, both as to private collections and public 
museums, teems with chests, chairs, and other articles approximating from the period 
of Louis XI., contrasting very forcibly with our own scanty relics of the same type 
and period. French framed chests were more elaborate in their joinery and adorn- 
ments than the English. The locks were often marvels of intricate workmanship, 
and throughout almost the whole ran the sinuous convolutions of the Flamboyant 
style, assisted by the linen panel. The French exhibited intersecting mouldings on 
the edges of framework (as reverted to elsewhere in this work), also a slight archmg 
over the panels, seldom seen in English furniture of th'is age. The tracery, too, 
which adorns French woodwork of the Flamboyant period, is often embellished with 
a curious diamond-shaped ornament, which, singularly enough, does not seem to have 
been employed in their stone carvings, and which ma)- be looked for in vain on any 
of our own national productions. 

French chests were more frequently buttressed than those made on this side of 
the Channel, where a great many receptacles reverted to the simple " coffer " or single- 
panel type. These buttresses were sometimes adorned with " paned " and spiral 
carvings of a very elegant description, into which presently began to creep the 
leafage of the superseding styles. 

A good many oak chairs of late fifteenth -century workmanship remain in France, 
most of them being intended for state occasions, and some very magnificent speci- 
mens may be seen in the French museums, adorned with buttresses, and carved 
on the panels with designs in the Flamboyant style. In Rouen Museum an arm-chair 
is preserved which is small and of a somewhat more unassuming type, but which has 



A HISTORY OF OAK FURNITURE 17 

the back perforated with late Flamboyant tracery. The type is a very uncommon 
one, and details are not wanting that indicate it to be a product of the Transition 
period at the junction of the centuries. The little dainty Gothic stool from Normandy 
now in our Victoria and Albert Museum is well known, and is a gem of its kind.* 
Its lines should be compared with the English specimens of approximately the same 
date, shown in Plates XXVI., XXVII., and XXVIII., and with the slightly later 
Italian example in the Victoria and Albert Museum. t 

The first breath of the Renaissance did not immediately revolutionise all the 
features of the late pointed style. Scores and hundreds of pieces exist which exhibit 
the distinctive peculiarities of the two epochs side by side with each other. The 
Cluny galleries possess many such specimens, where the synthesis, far from being 
inconsistent, presents a most pleasing and harmonious appearance. The Gothic died 
hard, and for a time it actually seemed uncertain which style would absorb the other. 
But the cult of classic pilasters, foliated scrolls, arabesques, caryatides, and profile 
masks eventually prevailed, and the Gothic went down, never to recover itself, though 
a rugged survival existed in Brittany till well into the seventeenth century. 

At first the old form of outline was adhered to in furniture as in ecclesiastical 
buildings ; armoires and credences underwent no change in structure, though what 
may be termed the "trimmings" differed considerably. One has only to look at a 
multitude of French churches, built or added to about the Transition period, to 
understand how difficult it was for the old designers to start their reformation of 
style on anything more elementary than the frills. To mention only a few examples, 
the churches of St. Eustache, in Paris, commenced in 1532 ; St. Jacques, Dieppe, built 
in the sixteenth century ; the chapels surrounding the choir of St. Pierre, added in 
1521 ; and the remarkable oak screen to a chapel on the north side of the choir at 
Evreux Cathedral, may suffice. In each of these exemplifications the prevailing 
outline is arranged upon Gothic principles, while the details in character belong 
essentially to the Italian Renaissance. Theoretically, such anomalies are not only 
wrong, but in very bad taste, yet the ensemble is generally rich, and the adaptation 
shows great inventive faculty. 

During the second half of the fifteentli and throughout the following century 
walnut-wood was much used by the French in the manufacture of furniture. Walnut 
is of a finer and smoother grain than oak, and is very desirable material for the work- 
ing of small and intricate details by the carver. Where sap-wood occurs, it is peculiarly 
liable to the attacks of the worm and decay consequent upon exposure to damp, but 
where the tree is sound, and pieces worked from such portions are properly kept, 
walnut will often retain its sharpness to a wonderful degree. It is no unusual thing 
to see French cabinets and chests made of walnut in which sections are so rotten as 
to be positively devoid of outline or detail, while adjacent surfaces remain uninjured 
and exhibit their pristine cutting as clearly as the scoop-marks on a Gruyere cheese. 



* Normandy oak stool. \o. 96S, 1897. 

t Italian walnut stool, given by Lady-^Mond. No, \V. 10, 1910. 



i8 A HISTORY OF OAK FURNITURE 

Numerous examples of pseudo-classic productions may be seen in the Cluny, 
Orleans, and other museums ; and, while studying the complete examples comprised 
in these collections, I have been forcibly struck with one thing — in furniture of the 
Gothic type the French, so far as beauty of outline was concerned, were able to hold 
their own with any country in the world, but when it came to their merely adopting 
Italian mfluence, pieces from the southern country were more subtle and exquisite 
as far as contour is concerned, be the carving what it may. If two similar pieces 
of furniture of the Southern French and Italian Renaissance respective!}', and quasi 
equal merit as regards execution, be closely compared together, perhaps no superficial 
superiority may be observed in either article; but view such pieces from a short 
distance, merely as a mass of contour, and the more refined outlines of the Italian 
production almost invariably assert themselves. 

About 1560 French furniture had assumed a floridity which was often over- 
elaborated, and which, though the sculpture was frequently highly finished, failed 
to satisfy as in their Italian prototj'pes. French design was apt to be overdone and 
unrestful after the middle of the century was passed, and even in the most superb 
productions of French craftsmen one often sighs for the introduction of a plain 
surface to contrast with and accentuate the lavish carving in high relief. Even 
the very elaborate and well-modelled walnut buffet, or sideboard, from the Soulages 
collection, now in the South Kensington Museum, shows a want of restraint, while 
its outline leaves much to be desired. This curious elaboration may be mainly 
traced to the influence of Jean Gougon, called the " French Phidias," and his attempts 
to establish a national school of sculpture and design. Gougon appears to have 
been born in 1515, when the Early Renaissance was in full swing, and we first hear 
of him as being engaged in embellishing and beautifying the cathedral of Rouen 
and the church of St. Maclou; grafting, in fact, on an older mode. Though his 
art was in a way founded on classic models, he in great measure discarded the Italian 
style and adopted an original treatment quite his own. He was employed by 
Henri II., and executed a vast amount of pediments, statues, and decorations, among 
his chief efforts being bas-reliefs upon the Hotel de Carnavalet, the Fontaina des 
Innocents, and a portion of the facade of the Louvre. Jean Gougon was suspected 
of being a Huguenot, and tradition has it that he was shot during the massacre of 
St. Bartholomew while engaged at work in the last-mentioned building. This oral 
legend, which is an exceedingly old and persistent one, has been controverted witli 
some success; but, at all events, we hear no more of Jean Gougon, and his influence 
gradually declined. The carvings upon French cabinets and chests of Gougon's time, 
both in high relief, were frequently executed with much delicacy, but the accompanying 
mouldings were heavy, and surfaces were clogged with an excess of ornament. Yet, 
with all its defects, this type of art is distinctly a national one, and deserves full 
recognition as a movement which proceeded from within. The decline of Gougon's 
influence in the last quarter of the sixteenth century marked a change in the style of 
furniture, the subject of which must be reserved for a future occasion. 



I'LAIE V. 




NOKMANDV STOOL. 



FIFTEENTH CENTLRV. VICTORIA AM) ALBERT MLSEUM. 




CHAPTER IV. 

Furniture of the Netherlands. 

I HE quantity of oak furniture which sprung from Holland in bygone 
times is a matter for wonder, but the majority of pieces which we are 
acquainted with date no earlier than the seventeenth century. Old oak 
furniture of that period is even now not unplentiful in the market- 
that is, except in Holland itself, where the greater part of the so-called antiques 
on sale are fabrications, many of them made and exploited by Frenchmen. The 
oak tree does not flourish in Holland, and therefore must have been brought down 
the Rhine from Germany in vast quantities for consumption in the land of bulbs 
and willows. The oak Dutch furniture is of a very fine grain, the medullary rays 
not being very distinctly marked. It is softer in substance than English oak, and 
is capable of being worked by the carver to a high degree of finish. Practically 
very little remains of Dutch furniture of any period preceding the Renaissance, but 
from what we know the natives of Holland, unlike their Flemish neighbours, did 
not take very kindly to Gothic forms in the arts. The custom of painting their 
furniture with floral emblems was lavishly carried out by the Dutch during the 
sixteenth and seventeenth centuries, but the effect produced by such embellishment 
was gaudy, and during the period of its first freshness must have been horribly 
distracting. The art is said to have been last practised at the little village of 
Hindeloopen, and to have finally died out during the nineteenth century. The old 
Dutch Masters were proverbially so faithful in their renderings of contemporary 
life, that we can study their pictures with the greatest profit when determining the 
shapes and peculiarities in their furniture. In an oil-painting of high finish by 
Gerrita Berck-Heyde, representing the interior of the church of St. Bavon, at Haarlem, 
and dated 1673, we may see a fine large specimen of an early sixteenth-century 
church coffer, possessing a plurality of locks. This coffer, which is painted a red 
colour, stands upon clamped feet. 

A Dutch or Flemish coffer, boldly carved with figures, is in the Soutli Kensington 
Museum, the subject on its front panel representing Esther and King Ahasuerus, and 
bearing this inscription: — 

"DE SICK GADES VOLCK VMME THO BRINGEN HADDE VOR 
GENAMEN ISTOM LESTEN SVIV." 

The figures measure some six inches each in height, and their costume shows 
here and there indications of modes of the sixteenth century, to which period 
the piece doubtless belongs. Portions of this coffer have been replaced, though it 

19 



20 A HISTORY OF OAK FURNITURE 

is probably in its original shape. The uprights are richly carved with pilasters and 
swags, and the whole effect is heavy. 

As a type of florid Dutch woodwork of the period, the ornate and yet refined 
choir-screen at Monnikendam Church, dating from 1562-63, should be studied with 
much greater advantage. Generally speaking, the outlines and projections of Dutch 
furniture are bold, while carved decoration, apart from figures, is executed in low 
relief, scrolls and arabesques being rendered in a manner distinct from those of 
any other nationality. During the period of the Dutch Renaissance, the doors of 
cabinets and presses mostly revolved on pin hinges, placed vertically, thus being 
invisible, and playing no part in the decoration of such pieces. One extremely 
probable reason that so few pieces of Dutch furniture remain of any date prior to 
the seventeenth century is the long and destructive struggle which was carried on 
against the Spaniards under Alva, and the devastation which was caused in this small 
but obstinate and intrepid country. No people have ever shown themselves fonder 
of inscribing the dates of erection upon their dwellings than the Dutch, a great 
number of houses yet remaining in Gronengen, Maastricht, Dordrecht, Delft, and 
other places which bear sixteenth-century dates, ranging from 1509 to 1 571. This 
custom was doubtless carried out also with regard to specimens of furniture, which, 
if they had remained in existence, would have afforded an invaluable guide at the 
present day. We know that after the " Spanish fury " had spent itself, and the 
deposition of Philip II. had been accompanied by the appointment of William the 
Silent as Prince of Holland and Zeeland in 1581, the departing Spaniards left 
many traces of their occupation behind them ; leather-backed and seated chairs, inlaid 
cabinets, and Jnspains, or window casements closed with shutters doubly-hinged, as 
well as other details, remained to show how deeply the southern race had left their 
mark upon the domestic life of the nation. 

With Dutch oak chairs of the seventeenth century we frequently find that the 
material employed was not so lavishly used as in English examples, and, contrary 
to the custom obtained in cabinets and cupboards, the leg and stretcher members 
are flatter and more restricted as to variety of outline. Table legs, however, exhibited 
enormous bulbs, though, contrary to the English custom, such bulbs were seldom 
decorated with carving other than a sunk band or a ridge passing round them. 

Cupboards often possessed very wide cornice mouldings, underneath which 
appeared an elaborately carved frieze, while the facades were supported by pilasters, 
the upper shafts of which were frequently ebonised, the lower portions, or " drums," 
being carved in ordinary oak. The doors were sometimes embellished with depressed 
arches, but more often exhibited small square panels, surrounded by wide mouldings, 
and bearing in their centre quasi-classic patterns very different from our English 
strap-work. A brother brush who visited Hoorn between forty and fifty years ago 
informs me that on market-day there he has seen on the "Plein," among the objects 
exposed for sale, numerous oak cupboards and presses of the seventeenth century, 
each one of which would have served as an example for students of design. All 
of these have now disappeared. Holland is swept clean, though a systematic trade 



A HISTORY OF OAK FURNITURE 21 

is still carried on in some places (notably at Marken), where the cottages are restocked 
again and again for the benefit of those visitors who wish to purchase souvenirs of 
the aborigines. 

Towards the end of the seventeenth century cupboards and coffers became heavy 
and lifeless in design. Such pieces are wanting in the fine lines which we usually 
associate with the Dutch Renaissance, and possess little attraction for the connoisseur. 

A great part of Flanders suffered under the same infliction of the " Spanish 
fury " which scourged the sister country, but resistance was less successful, and it 
may be that destruction was not so complete. At all events, we had till lately no 
lack of old Flemish furniture remaining of every date from the Gothic decline 
downwards. 

Difficulty is often experienced in discriminating French tracened chests from 
similar productions made by Flemish hands, though the former examples are generally 
superior and more satisfying to the connoisseur of old oak. The flambo)ant style 
was as popular in Flanders as it was in France, and many contemporary details 
were identical in these two countries which will never be found in English work, 
notably the intersection of mouldings, which is referred to in another part of this 
work, and the rounded-top framing over panels of late Gothic design. But Flemish 
work, to me at least, has often a somewhat disappointing effect, producing the same 
sort of feeling which one has on viewing the west front of Brussels Cathedral after 
lingering over the same aspects in matchless Amiens or Notre Dame de Paris. Not 
that this simile in any way refers to their respective dates. A fair number of 
"Flanders chests" still remains in our own country to attest the export industry of 
the Flemings during the second half of the fifteenth century. 

When the Renaissance had taken root, the Flemings seemed to apply themselves 
with extraordinary vigour to the production of florid specimens of the new style, 
and a vast quantity of productions, dating from the sixteenth and seventeenth centuries, 
exist, in which its peculiar features are pushed to their utmost limits. If heads 
carved on French and English woodwork verged upon the grotesque, those yielded 
by the Flemish were even more so. A species of heaviness, too, hung about their 
representations of the human figure, which is consistent with the powerful frames 
but somewhat sluggish temperament of the people themselves. Mouldings did not 
escape this ponderousness, and many a noble standing cupboard, which in the hands 
of the French or Dutch would have developed fine proportions, is marred by a want 
of symmetry in its component parts. Some good specimens of ancient Flemish 
furniture were to be seen before the great war at the Musee dArt Industriel Ancien in 
Brussels, and the Musee Plantin, Antwerp, where, in the dwelling portion, the presses, 
cupboards, and chests used by the Plantin famil)- during tlie sixteenth and seventeenth 
centuries remained in their original surroundings. 

Towards the end of the sixteenth and during the following century the Flemings 
were very partial to split-pendants and balusters applied to the stiles of tlieir 
furniture. Ornaments of applied strap, perforated in fretwork, were also fastened 
on the uprights, giving an appearance of richness, which is very pleasing. The 



22 A HISTORY OF OAK FURNITURE 

latter custom does not appear to have become fashionable in England, where such 
ornament was almost invariably carved from the solid. The doors of oak cabinets 
and cupboards were also quartered into geometrical panels, but the proportions and 
effect were distinctly different from similar methods of framing used by French or 
English craftsmen. Marquetry, as employed by the Flemings, was often overdone 
and in doubtful taste, the execution being mostly inferior to embellishments of the 
same nature executed by either French or Germans. 

The Flemish Baroque, or decadence of the Renaissance style, is perhaps as ugly 
and unmeaning as any declension in Europe. Effects were perpetrated which 
occasionally give one the impression of explosions in wood or unexpected accidents 
happening to the stage scenery. These, however, are for the most part to be found 
in churches, domestic furniture partaking more heavily in their appropriations of 
French and German characteristics, and exhibiting a combination of the baser methods 
of both nations. 




Plate VI. 




a 




a 




u. 




u. 




o 




u 


.^■ 










I 


%) 


z 




B 


■^ 


a: 


*^ 


Uh 


^J 




-CJ 




•^ 


z 


■^ 


< 


^ 


tfl 


t= 


w 


^ 






< 


o 


M 


,0 


a 


s 




CHAPTER V. 

Tyrolese Furniture. 

:X Tyrolese carving a peculiarity exists which can hardly fail to be noticed, 
even by novices. The decorative designs are made by outlined incisions, 
leaving the upper surface of the sculpture perfectly flat, and not moulded 
in any way. In fact, the objects represented may almost be termed 
silhouettes. Markings on scrolls, animals, mythical creatures and the like, are all 
indicated by slight incisions, modelled surfaces being avoided. The lower planes 
in the sculpture — the " field," if an heraldic term may be employed — are frequently 
coloured with vermilion, red ochre, and black, accentuating the flat, outlined objects, 
which are represented on the higher surfaces. The locks, lock-plates, and iron 
fittings generally, are very elaborate and decorative in character, displaying a pro- 
fusion of wrought foliation and scroll-work, generally backed with scarlet cloth, after 
the German manner, and of which traces may sometimes be found on existing 
specimens. 

The staple material of which Tyrolese furniture is constructed is fir; no doubt 
chiefly owing to the indigenous nature of that wood, but perhaps also in a small 
measure clue to the singular immunity which it possesses from the attacks of both 
moth and worm.* It is probably these latter qualities which have been instrumental 
in saving so many specimens of early Tyrolese work from the wreck of old estates 
which for a hundred years and more devastated the district. From what is left it 
may be safely assumed that the quantity of fine productions in the shape of furniture 
must have been immense, the thick forests which cover nearly half the country 
contributing easily to the manufacture of furniture by the craftsman. Another 
singularity in Tyrolese decoration is the survival of Gothic characteristics in pieces 
made in times when such features had been fairly abandoned by the rest of Europe. 
The chest belonging to Sir Coleridge Grove, which figures in the illustration on Plate 
LXVII., is an example of how pieces were produced in the seventeenth century bearing 
very elegant Gothic tracery, accompanied by debased arches and the undisputed dates 
of their construction. There is not a few of these anomalies about, and their apparent 
contradictions are not unfrequently puzzling to the amateur. Occasionally other 
enigmas arise in connection with specimens connected with this group. For purposes 
of cross reference the typical Tyrolese coffer, bearing an English inscription, now in 
the Victoria and Albert Museum, may be briefly noted here, though it is more fully 
debated on in the chapter entitled " English or Foreign ? ". 

Some of the early fruit of Tyrolese craftsmen is intensely original and artistic 

* Cedar and walnut were also occasionally used, the latter wood being mainly employed in 
the shape of inlay. 

23 



24 A HISTORY OF OAK FURNITURE 

in its imagining, composite pieces answering the several purposes of cupboard and 
settle, chest and dresser, being contrived in a way which could hardly have been 
excelled in any other country in Europe. Towards Bavaria the characteristic flat- 
surfaced carving began to be mingled with a more modulated contour, and especially 
in the neighbourhood of Munich the splendidly elaborate scroll and foliated work 
in high relief, so distinctive of the Teutonic sculptors, asserted itself. A combination 
of these two methods of decoration on one article often presents a very agreeable 
sense of relief. 

The Tyrol, in common with many other parts, may no longer be considered the 
happy hunting-ground of the collector, though examples of the National Art may 
still be purchased there, but at prices of an altitude out of all proportion with those 
of a decade or so ago. Indeed, some of the amounts at which these relics were 
acquired by museums and pioneer collectors would hardly purchase a good deal 
table with its ordinary accompaniments. But that is forty years or so ago, and 
nowadays the Tyrol is scoured as much as many other hunting-grounds in Europe. 

The ingenuity of the old Tyrolese craftsman in planning fixed furniture such as 
combined settles and dressers, or seats and cupboards, has been somewhat revived by 
so-called art furniture designers in our own country within recent years, but much 
has to be accomplislied before ancient specimens can be successfully competed with. 
The simple needs of the living-room have to be more seriously studied than a mere 
striving for effect. Once this desideratum is recognised, true picturesqueness may 
follow as a natural result. 





CHAPTER VI. 

Some Fresh Figured Examples. 

HAVE dealt elsewhere with the group of knightly examples, which, for 
purposes of classification, I have dubbed with the imaginative title of 
" Tilting Coffers." * The description of this very rare type included 
specimens at Harty, Kent; Southwold, Suffolk; York Minster; Ypres, 
Belgium; and the Victoria and Albert and Cluny Museums. Since the publication 
of the works dealing with this subject, one or two other prototypes of a similar kind 
have come to light, which are fully as interesting as any hitherto described. The 
most important of the fresh examples is a coffer-front now at New College, Oxford, 
bearing a somewhat confused but remarkable carving, evidently intended for the 
battle of Courtrai, which occurred between the Flemings and the French in the 
year 1302. The banners of the various Flemish guilds are depicted, as well as 
various incidents in the combat, and though somewhat clumsy in execution, this piece 
is replete in historical interest of a definite kind. The arms of Pierre Conine, one 
of the Flemish leaders, are shown on one of the banners, and, it is said, are to be 
found on no other records except on a seal in the archives of Bruges. This coffer- 
front was debated upon at some length at the Society of Antiquaries on 19th March, 
1914, when an exceedingly able and descriptive paper was read by Mr. C. ffoulkes, 
Keeper of the Tower Armoury, who made out a very good case for a curious weapon 
with which the men gathered under the Flemish banners are depicted as being armed, 
and which he identified with the " Godendag," or " planqon a picot," a species 
of club topped with a spike, which is mentioned by the chronicler Guiart as being 
in use about this date. As regards the general mode of decoration, a very extra- 
ordinary survival of an earlier epoch is discernible in the crowding of the figures in 
a sort of processionary movement into longitudinal compartments placed one above 
the other on the same panel. The Franks casket, of Anglo-Saxon workmanship, 
in the British Museum, is treated in a similar manner, and there are many parallelisms 
between the methodical treatments of these widely dissociated pieces. 

Nothing is actually known about the history of this coffer-front except that it 
formed an item among some goods taken in lieu of rent of an Oxfordshire farm 
some years ago. But circumstantial history in such absorbing relics is seldom silent, 
and some attempt has been made to connect this coffer-front with the great poet 
Chaucer. The " father of English literature," as he is termed, married a Flemish 
lady, and owned a demesne in the vicinity of the farm from whence the coffer-front 



* Ancient Coffers and Cupboards (Methuen, 1902); Old Oak Furniture (Methuen). 

25 



26 A HISTORY OF OAK FURNITURE 

came, and it is also known that Chaucer conducted secret-service work in Flanders 
in the years 1376-7. But proof of any sort of the association is wanting. The 
costume and equipment of the figures more nearly approximate to the date of the 
battle of Courtrai than to Chaucer's sixty years or so, ending with 1400; but while 
there is no reason that Chaucer should not have possessed a piece of furniture some 
seventy or more years old, it is hardly feasible to suppose that so heavy and militant 
a piece w-ould have taken the form of a dower chest to the poet's wife. The conjecture 
raised is certainly interesting, but it is purely presumptive, and lacks any real 
supporting evidence. That the subjects depicted were intended for incidents 
connected with the battle of Courtrai, where the obstinate valour of the Flemish 
burghers repulsed the fiery charges of the chivalry of France, there can be hardly 
any doubt, but beyond this one can scarcely venture. It is to be hoped that discussion 
on this relic may help to bring into prominence other pieces of a like nature. 

The second specimen in this class which appears to have escaped any serious 
notice is a carved and coloured wooden panel which hangs on the north wall of 
St. George's Church, Norwich. The subject represented in the sculpture is the 
patron saint's encounter with the dragon, and though later in date than the 
better known specimens at York, Ypres, and South Kensington, has some of the 
features of the coffers bearing the same theme. This panel evidently formed part 
of the muniment coffer of the very church in which it now remains. At present it 
is in a very bad position for examination. 

A coffer-front of the fifteenth century, exhibiting extraordinary characteristics, is 
possessed by Colonel Walter Horsley, of Cranbrook, Kent, which may well be 
included among the figured examples. The tracery which adorns this interesting 
fragment is unmistakably English, but apparently the craftsman who executed the 
carving was in such a hurry to commence his task that he quite neglected to set out 
the design beforehand. The piece belongs to a time and type of which we have none 
too many examples remaining, but what specially enhances its interest is the fact that 
above the traceried arches appears the representation of a female head, gardant, and 
bearing the enormous double-horned head-dress so typical of the period. The lady's 
head (which is uncomplimentary to the sex) is accompanied by coarse chimeras, while 
on the sinister upright is carved an unequivocal devil of hideous aspect. The opposite 
upright is decorated with a shield bearing a saltier engrailed, on the dexter side a 
martlet. On that end of the panel nearest the shield appears a demon's head, 
the tracery of the windows being " broken," or interrupted, to admit the insertion 
of the solitary black letter t- The whole design is in a weird and fantastic vein, 
the figures being of a morbid nature reminiscent of the old " Dooms." Read the riddle 
who can of this cryptic and singular relic. Its history is unknown, beyond the 
certainty that it has been in its present abode for at least forty years. 

Very different as regards quality are some of the pieces in the collection of wood 
carvings formed by the Rev. Philip Nelson, of Calderstones, Liverpool. These are 
partly heraldic in character, and indicate such a high state of art that it is a matter 
for regret so little remains. In particular, one magnificent fragment of panelling 



A HISTORY OF OAK FURNITURE 27 

which was discovered during some repairs at Docker Hall, Lancashire, is the finest 
of its kind, the knightly figures, armed in harness of the fifteenth century, being 
designed with exceptional skill and fidelity. We frequently find circumstantial 
attributions tacked on to roughly executed pieces — such as the armoire at South Ken- 
sington, or the Courtrai panel — based, perhaps, on little more than their discovery near 
some ancient manor, while here is a remnant of the highest class obviously made for 
a personage of great distinction about which no conjecture seems to have been hazarded. 
The breaking up of a piece of antique furniture almost always destroys much 
evidence as to its origin. The carving of human figures, whether French, English, 
or otherwise, may possess points of similarity, but the reaving of a panel from its 
framing can have no other effect than to get rid of any proof which its attendant 
mouldings might afford, especially in the case of objects dating from the fifteenth 
and early sixteenth centuries. It may be accepted as a truism that the English 
designer or craftsman invariably abstained from the intersection of mouldings on 
the corners of framed panels which characterises so much of the continental work. 
Elimination of contributory evidence of the mouldings has, I feel sure, been responsible 
for many errors of judgment in descriptions attached to objects deposited in our 
museums. How important the foregoing thesis is may be at once grasped when the 
mouldings are not attached to the framing, but carved on the surface of the panel, 
and, consequently, have not been destroyed when the breaking-up process occurred. 
Even in the latter instance such manifestations are not always fully weighed. Two 
little credence panels which have recently been acquired for the Victoria and Albert 
Museum are a case in point. These pieces, which date from the junction of the 
fifteenth and sixteenth centuries, have, superficially, a Flemish twist about their 
sculpture, though perhaps not enough to give absolute denial to the description which 
is attached to them: "English, about 1500. W. 5-6, 191 1." If, however, the 
intersection of the mouldings on these panels had been properly regarded, I venture 
to think that this decision would have been a different one. The point is one well 
known in the architectural profession, but does not appear to have been sufficiently 
considered by many critics of antique oak. Here, again, the same difficulty may 
present itself as to the precise meaning of the word " foreign," but I may say that 
I know of no instance in which pieces of oak furniture exhibiting these intersecting 
members can, on close examination, be rightly referred to as of English origin, either 
as regards design or execution. 




CHAPTER VH. 
Secret Receptacles and Treasure Chests. 

FAIR amount of ancient church coffers still exist which retain their 
original hidhig-places in various forms, sometimes in the shape of a 
false bottom to the body of the coffer itself, but more often fitted beneath 
the little tray which is almost invariably attached to one of its ends. An 
example of the first type remains in the parvise of the parish church at Newport, 
Essex — an exceedingly rare and beautiful specimen of ancient English furniture, 
which is more fully described in another of my works; while instances of the 
secret tray-bottom may be found in coffers existing in the churches of Long Stanton 
.St. Michael, Cambs. ; Stoke d'Abernon, Surrey; and Rogate and Bosham, Sussex. 
That the hiding-places in these ancient coffers were sometimes hastily emptied, there 
is every reason to believe, and we have evidence of this in at least one case in Southern 
England. The coffer at Bosham, just mentioned, is a plain, heavy production, 
unornamented except for dwarf pilasters at the base of its huge uprights, and some 
chamfering on the lid flanges and inner tray lid. A few years ago a thorough 
examination of this coffer was instituted by the incumbent, with the result that a 
silver Waterfoid halfpenny, dating from the reign of Edward I., was discovered in 
the secret well beneath the tray. Though the device must have been fairly well 
known to the craftsmen of the thirteenth century, it was sufficiently ingenious and 
mute to have safeguarded its secret during several dynasties. 

But, nevertheless, repetition must have ventilated this secret more or less, for we 
find that changes occurred in the fashion and devising of concealed receptacles, 
which proves that in some contemporary cases discovery must have been suspected 
or else taken place. All the coffers just mentioned belong to the thirteenth century; 
after that the tray-well was discarded, only to reappear in a few examples made in 
Elizabethan times.* It is evident that some novel device or custom appealed to the 
patron or cofferer, and the old contrivance not only became obsolete, but entirely 
disused and forgotten. A few specimens of later improvisation may be mentioned. 

In the early part of 191 3 a Tudor oak buffet was sold at Messrs. Christie, 
Manson & Woods', which contained two secret receptacles enclosed by sliding panels 
in the front. This buffet was a very elaborate specimen of workmanship, and fetched 
the large sum of ;£^840 under the hammer, despite certain indications that seemed 
to hint attentions at the hand of the restorer. 

A seventeenth-century French armoire known to the writer possesses a very 
artful hiding-place for money in the foot of each of its uprights — a very unusual 
position for such receptacles. 

It is not till we reach the borders of the seventeenth century that court cupboards 



* The late Mr. Ernest Crofts, R.A., possessed several family pieces of oak dating from the commence- 
ment of the seventeenth century which retained hiding-places of this description. 

28 



Plate VII. 




jfc-.- ..-^r«»>»:.f^— -.. -.^- --— -^ .^ -^^ „^ ,^ -fm-'jnr. T 




WALNUTWOOD CABINET. SECOND HALF OF SIXTEENTH CENTURY. FRENCH (LYONS). 
Victoria and Albert Museum. 



A HISTORY OF OAK FURNITURE 29 

attain the little secret ledge or shelf over their top cupboards which may be found 
in so many English examples. To those who know not this simple peculiarity, it 
may be briefly explained that it generally exists in recessed pieces of tlie court 
cupboard class, immediately between the cornice moulding and the niche underneath, 
and may be easily found by opening one of the small upper doors in the recess and 
placing the hand upwards in the crevice formed by the canopy. A somewhat well- 
known and obvious place of concealment, even at the time when such pieces were in 
fashion, it is little more than a hidden shelf, and could hardly have been used 
except for temporary purposes of secretion ; \'et it nowadays often escapes notice, 
and since the old style of court cupboard has ceased to be made, except by fakers 
and imitators, this humble hiding-place often remains undiscovered for an indefinite 
time, just as the secret wells in the coffers of the thirteenth century rested perdu once 
the fashion of their making was over. 

But such " secrets," as they were sometimes termed, could hardly have been 
intended to receive treasure of any bulk, or objects which for ordinary safety would 
frankly be locked up in the strong box of the time. The iron-bound treasure-chests 
such as we usually associate with the Great Armada or the safe-keeping of early 
muniments, are merely cumbrous, unornamental affairs, and proclaim boldly without 
disguise that they defy any attempt at unlawful opening. A great many of the latter 
specimens, both wood and iron, possess false or dummy lock-plates, intended to 
deceive the would-be thief, though this custom was so well known at the time of 
their construction that it is hard to believe that many intellects could have been 
duped by so shallow a pretence. 

Wonderful stories are told of these so-called Armada relics, many of them of 
quite a circumstantial nature, and by far the greater part absolutely without founda- 
tion. The screen and gallery in Middle Temple Hall, a typical piece of English 
carving dating from 1575, has similarly been asserted to have come from one of the 
captured Spanish galleons, and this absurd and preposterous fallacy is still believed 
and spread by many ignorant people. Most of the iron-sheathed strong boxes to 
be found in cur municipal museums have legends of this description attaching to them. 
Such fables are as numerous as the stories about Cromwell, and certainly on the 
whole as baseless. 

Amongst the wonderful collection of objects connected with the City of London, 
gathered together in the Guildhall Museum, are four typical specimens of the iron 
strong-box of the sixteenth and seventeenth centuries, about which a certain amount 
of record is known. The first three are the ancient muniment-coffers of the Ship- 
wrights', the Feltmakers', and the Plumbers' Companies ; the fourth formerly filling 
the same purpose for the Church of St. Vedast, in Foster Lane, a building which 
was only partially demolished by the Great Fire of 1666. All these coffers have 
complicated locks inside their lids, provided with from ten to thirteen bolts, as 
well as hasps and padlocks on their exteriors. The coffer which formerly belonged 
to the Worshipful Company of Shipwrights is furthermore fitted on one side with 
an inner box, a sort of survival of the Gothic traj-, affixed, however, to the bottom 



30 A HISTORY OF OAK FURNITURE 

and not to the upper part, as in earlier examples. All these three municipal guild- 
chests were presented to the Museum by their respective companies. 

Examples of treasure-chests of wood bound or sheathed in iron are numerous, 
and may be seen at Salisbury and Wells Cathedrals; the West Gate, Winchester; the 
Town Hall, Fordwich, Kent ; the Rolls Museum in Chancery Lane, and a hundred 
other places. The various dates of their construction are often obscure and difficult 
to detect, as some trifling peculiarity in construction of fitting is frequently the only 
detail which affords any light upon the subject. We have, however, some evidence 
of the approximate genesis of the Winchester example, as it is known to have been 
made to replace an older one which was broken into by thieves in 1590, when the 
city seals and plate and " four score pounds " were carried off. The ringed handles 
for transport at each end of the coffer are noticeable. See Plate XLII. 

We have an interesting record of the sums disbursed for the construction of 
the iron-bound coffer which exists in Ashburton Church, Devonshire. In the year 
1482-3, "John Soper, for sawing — feet of timber," was paid two shillings and four- 
pence, while " John Clyff, for making one chest," received the sum of 6d. The said 
box was apparently not considered strong enough, for six years later a " Mr. Half- 
hyde " received 13s. io|d. "for iron and making the same for binding the great 
chest," while a locksmith was remunerated with the sum of 5s. gd. for making the 
locks and keys. These several items, which appear to be singularly disproportionate, 
are recorded in the churchwardens' accounts, which have luckily been preserved. 

Instances of the discovery of treasure-chests containing their original contents 
intact are exceedingly rare in our own country, but they still occur occasionally, and 
always in totally unexpected places. As an example of how treasure may be quite 
lost sight of when it remains disused, the strange discoveries in Rochester may be 
quoted. The old treasury attached to the cathedral lately underwent a very neces- 
sary process of restoration, as the room had been practically disused for a long period. 
According to a writer in The Times of May 4th, 1914, " few people have been aware 
of the existence of the old room, part of the original Norman structure, or of the 
chief of the treasures which it contains. Only at intervals of years, so far as is 
known, have human feet trod the winding staircase which leads to the apartment. 
No one really seems to have cared what was contained in the great sixteenth-century 
chest, which must have been built inside the room, for there is not, and seemingly 
has never been, a door large enough to admit of its being carried through in its 
entirety.* If there ever was a list of the cathedral plate it has long been lost and 
forgotten, and some of the almost priceless pieces of silver which have been in use 
in the cathedral were commonly supposed to be of brass. The old staircase had 
become almost unclimbable, and the room itself was choked with the accumulated 
dust of, literally, centuries." 

* The custom of building large coffers inside the treasuries and parvises of ecclesiastical buildings 
was a common one, the main reason probably being that in the event of failure on the part of thieves to 
break open such receptacles, it would be impossible to transport them bodily away through openings too 
small to allow them to pass. The painted coffer in Newport Church, Essex, is an example. — Author. 



A HISTORY OF OAK FURNITURE 31 

The result of this much-desired spring cleaning was the bringing to light a 
quantity of long-forgotten plate of the most valuable description, chief among which 
were two beautiful dishes [ciboria) of the date 1530-1533. They are 5 inches high 
and 9 inches wide, and have between them one cover. It is conjectured that there 
was originally but one, which was transferable to either of the vessels. These relics, 
the discovery of which provided quite a sensation, are now exhibited in the renovated 
treasury among other archaeological valuables appertaining to the cathedral. 

Accounts of discoveries of a somewhat similar nature to the last occasionally 
reach us from the Continent, and one or two startling examples are given here. 

In 1907, while excavating in the streets of Madrid, some labourers unearthed an 
old chest of walnut wood, which, on being opened, was found to contain 300 doubloons, 
valued at about £^1,400 of our money at the time. 

In 1912 a discovery of exceptional importance was made in the village of 
Malaia Pereshtchepina, in the Province of Poltava, in south-west Russia. Some farm 
labourers, while digging in a field, uncovered a decayed coffer, which was found to 
contain a quantity of dishes, utensils, and ornaments of gold and silver, as well as 
some 450 gold and 50 silver coins, some of the latter dating back as far as the fourth 
century A.D. Owing to the fact that the coins and plates were alike considerably dis- 
coloured, the labourers, possibly because they could not believe in tlie existence of 
so great amount of treasure, persisted in regarding the ornaments and utensils as 
being of lead and brass, and divided their spoil between them, retaining it in their 
cottages. At length tlie news of this discovery got about, and on examination being 
made, the contents of the chest was found to include the following objects : —a large 
silver dish of very ancient origin, damascened in gold, and bearing a Latin inscrip- 
tion showing it to be in the possession of a bishop living in the eleventh century; 
a similar dish embossed with the image of a Persian monarch living in the fourth 
century ; eleven solid gold cups of Persian make, and a quantity of bracelets and 
ornaments, as well as several silver cups of Byzantine workmanship. The collection 
was exhibited in its entirety at the State Bank, and it was estimated that the aggre- 
gate value amounted to the astonishing sum of ;^ioo,ooo. An announcement was made 
that the owner of the land on which the treasure was found and the labourers who 
made the actual discovery would divide the value of the treasure between them — 
an exceedingly unwise proceeding, but probably more or less very problematical of 
accomplishment. 

Such finds of equipped treasure-chests may be excessively rare, but we arc, 
nevertheless, occasionally startled in our own country by the bringing to light from 
some long-disused and forgotten receptacle of objects hardly less precious. As an 
addition to our national relics, a discovery made at Colchester in 1907 may be classed 
among the most interesting made in recent years. The church of St. Giles, memor- 
able as being the burial-place of the two Royalist supporters, Sir Charles Lucas and 
Sir George Lisle, who were shot by order of General Fairfax after the siege of 
Colchester, was during the year mentioned undergoing a thorough restoration. An 
old oak chest in the vestry, which seems to have remained in a neglected condition, 



32 A HISTORY OF OAK FURNITURE 

was opened and emptied, with the result that there was found a purple cloth in 
excellent preservation bearing the insignia of the Lucas family and the date 1617. 
Whether this once formed the pall of the family mentioned, or whether its purpose 
had been the draping of the altar in the Lucas Chapel on the south side of the 
church, is doubtful ; but it seems likely that the relic had been secreted under the 
mass of documents with which the chest was filled in order to escape unwelcome 
notice and possible destruction. Putting a different construction on the term 
" treasure " than that employed in the cases previously mentioned, this discovery 
would be somewhat hard to beat. 

In days when property was insecure, and the only recognised bankers were the 
goldsmiths, many very ingenious devices were adopted in articles of domestic use. 
These were perpetuated through the decadent period of oak, the conservative instincts 
of ownership often resulting in a continuity which has descended to the present day. 
The annexed instances (latter-day discoveries culled at random) were perhaps made 
from no very early depositories, but the hiding-places kept their secrets indifferently 
well, though they must have possessed scores, and most likely hundreds, of duplicates. 

"A story which will set antique dealers examining all their stock for secret 
drawers comes from Cardiff. Mr. Muller, an antique dealer of that city, bought a 
bureau at a sale last November for £1. While Mrs. Muller was examining the bureau 
on Wednesday, she accidentally touched a secret button, which opened a hidden 
drawer, and in that drawer she found securities of the value of ;^i,000, in the name 
of David Morris Lewis. A daughter of this Mr. Lewis has been traced by the police, 
and it was explained that when Mr. Lewis died, six years ago, his family were 
surprised to find how little property he had left." — Daily Express, i8th April, 1913. 

At Birmingham, in 1908, the breaking up of an old family chest for firewood 
led to the discovery of a false bottom containing one hundred perfect spade guineas, 
worth some 25s. each. 

A rummage sale was held on the rectory lawn at Beardmore, Hants., in 191 o, 
when an old writing-desk changed hands for the sum of eighteenpence. Owing to 
rough handling a secret drawer subsequently fell out, and with it thirty gold coins — 
guineas and half-guineas of the reign of George III. 

Announcements such as these crop up almost every month, and, alas ! their publi- 
cation not infrequently results in much furniture smashing by ignorant persons in 
their desire to acquire a proportion of unearned increment. 

There are also many instances of coffers, chests, and cupboards which, having 
no secret hiding-places, yet possess some ingenious contrivance for concealing the 
locks or key-holes which secure their contents. A very customary and favourite 
expedient adopted by the Dutch and Flemish craftsmen may be specially described. 
In cabinets and cupboards possessing folding doors, the fa<^ades are often enriched 
with moulded pilasters of a classic character, the centre one of which is attached 
close to the edge of the overlapping door. The upper portion of the shaft of this 
column was made to slide horizontally for an inch or so, thus bringing the keyhole 
into view. A simple contrivance, and plentiful enough at one time, it may be that 



A HISTORY OF OAK FURNITURE 33 

the craftsman would not allow the lock to interfere with his original design, for 
the amount of secrecy involved could obviously have been but moderate, and the 
device must have been constantly liable to discovery owing to the recurrence of 
surface polishing, which was so religiously kept up in the Low Countries. The ornate 
Dutch cabinet illustrated on Plate LVI. is fitted with this mechanism. 

An English device of somewhat similar character, but of greater scarcity, is 
fully described in the chapter on " Furniture bearing Dates." 

A fairly recent sensation in the way of discoveries is that of the register of the 
Black Prince, which was unearthed from an old iron box in the keeping of a firm of 
solicitors of very old standing. The following interesting account of it appeared 
in the Daily Mail, May, 1914. 

The box "remained till 1878 in a vault of the firm in Bloomsbury. Then it 
was transferred to a Holborn street, where, after crossing the road, it remained till 

1913- 

"A responsible clerk who entered the service of the firm in 1863 was told by 
his predecessor, when he first took over his duties, that there was a ' diary ' of the 
Black Prince in a large oak chest, and the head of the firm occasionally used to 
refer to it by this name; but nothing was ever done in the way of examining the 
volume. 

" It was only last year, when the premises were finally transferred to the Inns 
of Court district, that the new head of the firm, being told of the existence of the 
' diary,' determined to have the large oak chest (some 6 feet by 3 feet) and various 
other receptacles for documents, including two rusty sheet-iron boxes, overhauled. 

'A number of interesting documents, including a fourteenth-century law book 
and a chancellery roll of Fountains Abbey, were found in the oak chest. In one of 
the two partitions of the old sheet-iron box, so rusty that the bottom was almost 
falling out, was lying the register of the Black Prince covered by a bundle of deeds. 
This was in the last days of April of last year. 

" The volume was soon afterwards submitted to one of the foremost experts in 
the country, who at once recognised its authenticity. With the owners' permission he 
submitted it to Sir Henry Maxwell-Lyte, Deputy-Keeper of the Record Office, who 
declared it to be the missing volume of a series of three registers of 1350 to 1365, 
the other two of which are in the possession of the Record Office, the one relating to 
Chester properties, the other to Cornish properties of the Prince. 

" The volume then returned to the possession of its owner. . . . Whether the 
volume will pass into the hands of a private collector or of the State the future 
will show." 

Among the items included in the volume the following are of special interest 
to students of the history of furniture: — 

" ' Robert Joyner ' received 60 shillings for three chairs, and for a leather-bound 
coffer 13 shillings and four pence." 

Of all discoveries in the shape of historical literature the last-mentioned is surely 
one of the most astonishing made within recent years. 

D 




CHAPTER VIII. 
Furniture Bearing Dates. 

JT would be difficult to classify and impossible to enumerate even 
approximately a tithe of the manifold examples of dated furniture 
remaining in this country. Private collections and noble and manorial 
residences abound in them, sales forced by death or other contingencies 
occasionally bringing to light treasures well-nigh as hidden as the Gothic crypts 
beneath our London soil. References, to be of any use to the multitude, can only 
be nmde to certain typical and accessible specimens, which may be studied freely, 
and their diverse dates compared with the structure and decoration of the pieces 
which exhibit them. 

It must be borne in mind that the terms " early " and " late " are comparative, 
and what would be an early dated chest would probably belong to no earlier age 
than that of Philip and Mary or Elizabeth. Of thirteenth-century coffers we possess 
a plenitude, but I am not aware that any of them bears the veritable year of 
construction upon it. On the other hand, the majority of dates actually appearing 
on furniture belong to the seventeenth century, any previous to the last quarter of 
the sixteenth being of the greatest rarity. 

One of the earliest dated examples of furniture in our own country is the coffer 
in Shanklin Church, Isle of Wight. This beautiful instance of Early Renaissance 
art is carved on its front with the letters " T. S.," formed out of the elaborate 
scroll-work so typical of the reigns of Henry VII. and Henry VIII. Running round 
these letters is a band, bearing the inscription : — 

"DOMINUS THOMAS SILKSTED PRIOR. ANNO DNI 1519." 

Underneath the lock appear the arms of the See of Winchester, of which Thomas 
Silksted was prior from 149S till his death in 1524. The Lady Chapel at Winchester 
Cathedral (the latest portion of that edifice) was completed by him after having 
been commenced by Prior Hunton, his predecessor. It is seldom, indeed, that the 
origin of a movable piece of furniture is so clearly indicated by its carved decoration 
as in the case of this fine coffer. 

Various conjectures have been made as to the reasons why so many Italian 
secretaires and nests of drawers should have been placed on obviously English 
stands. I think the riddle is not very difficult to solve. These products of Italy 
were in their own country mostly placed on side-tables, which, for some reason, may 
not have accorded with English taste or convenience, the result being that it is not 
uncommon in our land to find superstructures of distinctly Italian make associated 
with contemporary English stands. In our illustration (Plate XLIV.) the secretaire, 

34 



A HISTORY OF OAK FURNITURE 35 

which is made of camphor or cypress wood, is dated in two places 1594, while the under- 
stand is decorated with typical Elizabethan strap-work, carved in English oak. 
A curious feature may be noticed in the inscription, running round three sides 
of the front panel, which indicates that the craftsman who produced this piece must 
have been somewhat casual in the setting out of his lettering. The inscription reads 
thus : — 

DEVSINNOMINETVOSALVVEMEFACEINVIRTVTETVAIVDICAME 

At first sight, the legend appears to consist of a chaotic collection of letters, the 

order of which becomes apparent upon careful study. It will be noticed that the 

sign of abbreviation over the V should have been over the ME, the T of the ET 

being omitted, and that the carver miscalculated his distance to such an extent that 

he actually found it necessary to enclose an I and an A within two other letters at 

the end of the inscription in an exceedingly quaint way. The ME which the 

inscription finished with has been partly cut away at some time or another, apparently 

to make the falling lid fit. Contrast between Italian and English workmanship is 

very noticeable in this international product, the chip-carving of the Southerns being 

essentially lighter and more classic than the sculpture on the beautiful Elizabethan 

oak section. 

A very beautiful English court cupboard of the time of James I., which has 

recently been acquired by the autliorities of the Victoria and Albert Museum (W. 32), 

is of oak, carved and inlaid, bearing on its centre panel, underneath the canopy, the 

. . . A . 

initials „' at while the small doors to right and left of this panel are inlaid with 

the date "ANNO — 1610." No locks are visible on the upper tier, but on each side 
of the centre panel, separating it from the doors, is a heavily-moulded pilaster, the 
upper part of which is capable of being raised, when the keyhole is at once disclosed 
to view. This contrivance is a rare variation of the Dutch peculiarity mentioned 
in the chapter on " Secret Receptacles." The upper doors of this valuable and 
interesting cupboard are inlaid with geometrical patterns, and the frieze is carved 
with elaborate " jewel moulding," the lower doors being plainly panelled. Though 
missing its " cock's-head " hinges on the under story, this exhibit is in an exceptionally 
fine state of preservation, and is one of the best and most typical of its kind. 

At the time of writing, a very good Jacobean court cupboard is at the " Cheshire 
Cheese" in Fleet Street. This piece, which is of plain and Puritanical aspect, exhibits 
the inscription — 

T 
G D 
1626 
which seems somewhat to antedate its accepted type. 

Tables of the Elizabethan and Jacobean periods were frequently carved with 
the dates of their construction, the figures being usually distributed on two of the 
uprights and separated by the stretchers. A great number of these tables remain 
all over England, located mainly in churches, where they have been used for purposes 



36 A HISTORY OF OAK FURNITURE 

of Holy Communion since Reformation times, but also to be found occasionally in 
manor halls and farmhouses of the better class. The fine jewel-moulded specimen 
in Dinton Church, Bucks., dated 1606, described in my work on Old Oak Furniture, 
is a specific instance of this type.* Though such dates, if present, were almost 
invariably carved on the uprights of these tables and not on the stretchers, they were 
not always distributed on different uprights, as in the Dinton example, but were 
sometimes incised on one alone, with, occasionally, tlie maker's mark or initials branded 
beside them. 

A genuine trencher-table belonging to the Royal School of Needlework was 
among the objects on loan at the Exhibition of Ancient Furniture opened in 1914 
at the Geffrye Almshouses in Kingsland Road. This table was probably originally 
intended for an altar or side-table, as the front rail alone was carved. The legs were 
rather heavy in character, and on one of them appeared the genuine date, 1648, together 
with the maker's mark, " T H." 

Another piece of very great interest included in the same exhibition was a small 
oak desk, carved on all four sides with geometrical whorles, and on the lid with 
the arms of Oliver Cromwell, accompanied by the date 1659. The carving was small 
and somewhat Dutch in character, and had the appearance of being contemporary; 
the interior had been fitted with drawers, only one of which remained, bearing the 
inlaid inscription, "IS 1779." According to tradition, this desk was given by the 
Protector to Bernard Gomme, the engineer of Tilbury Forts, in the seventeenth century, 
its modern possessor being the late Sir Laurence Gomme. 

A curious little Flemish coffer in the Victoria and Albert Museum (410 — 19! i), 
shaped something in the form of a housemaid's box, is dated in iron figures 1556, 
accompanied by the initials "D. P. D." The wood is also clasped with ornamental 
bands of iron, and the lid is formed out of a very massive linen-panel, which, from 
its thickness, mig'ht have been removed from some church door. The type is an 
unusual one. 

Many boxes in English country churches exhibit carved inscriptions, accompanied 
by dates, though these are mostly pieces somewhat late in character. 

At Combs, Suffolk, is an oak-panelled chest inscribed with the comparatively 
early date 1599, while in the adjoining county of Norfolk a chest dated 1632 exists 
in the parish church of Great Snoring. 

A very curious and interesting chest, dating from the first half of the seventeenth 
century, is that remaining in the parish church of Norton St. Philip, Somerset. This 
relic bears on its front the somewhat involved inscription: — 

THOMAS JOHN . ME 

16 OjTENT REFELD 34 

CHURCH WARDENS. 



* The Dinton table'has been painted an abominable buff colour, through which the traces of inlay 
can be distinctly seen. 



A HISTORY OF OAK FURNITURE 37 

Local taste ( ! ) has of recent years mounted this chest upon legs, by which its 
appearance is by no means improved. 

In the parish church of Blundeston, Suffolk (famed as the Blunderstone of 
David Copperfield), is an oak chest of a latish type, scored with a crude pattern on 
its stiles and panels. This piece, compared with the beautiful poppy-heads which 
ornament the bench ends in tlie same church, is indeed a poor specimen, but a certain 
amount of interest attaches to it inasmuch that inside the lid appears the date 1640, 
accompanied by the initials of two people, most probably those of the churchwardens 
in office at the genesis of its use. This is a striking instance of the difficulty of 
assigning even fairly definite dates to some pieces of antique oak, as before seeing 
the inscription (an undoubtedly genuine one) nineteen authorities out of twenty would 
mentally "place" the piece as being considerably later in the seventeenth century. 

What is most probably one of the largest boxes in England must be included 
in the present section. It exists in the parish church of Scarcliffe, near Mansfield, 
Derbyshire, and is an enormous production, measuring some ten feet in lengtli, the 
lid bearing the incised inscription, " F. H. 1671."* Derbyshire is singularly rich 
in the possession of chests and coffers, even when those at Haddon are not considered. 
A specimen at Chelmorton in this county is carved with the somewhat pompous 
inscription : — 

RALPH BUXTON OF FLAGG GAVE THIS 1630. 

Among iron-bound church coffers exhibiting dates one very remarkable London 
example may be specially mentioned, which, though easily accessible, is not as well 
known or studied as it should be. This relic is the muniment coffer of the church 
of St. Giles-in-the-Fields, and though passed every day by connoisseurs and lovers 
of such rarities, remains almost unknown and unobserved. The coffer, which stands 
upon curious clawed feet, is bound with bands of iron, alternatively plain or perforated 
with strapwork decoration in the Carolean style. The date 1630, with the initials 
of the contemporary trustees, as well as the Royal Supporters and various coats of 
arms, appear on the front and lid of this coffer, the ornamentation being ingeniously 
carried out in wrought iron. St. Giles's Church was erected between the years 1720 
and 1734, replacing an older structure of red brick, which was consecrated by Laud 
in 1623, and the coffer just mentioned is evidently a survival from the earlier building. 
It is singular that Metropolitan histories, none of which forget to mention such 
circumstances as the association of the church with the Tyburn Tree journey or the 
tomb of Richard Penderel in its churchyard, should so persistently ignore the existence 
of this valuable and interesting relic, while often remarking upon the paucity of 
interest within St. Giles's walls. 

Of other dated specimens appertaining to church furniture, the poor-box in 
Dovercourt Church, Essex, should be noted. A ponderous lump of oak, strongly 



* The largest is a thirteenth-century coffer at Westminster Abbey, which measures close upon 
thirteen feet long. 



38 A HISTORY OF OAK FURNITURE 

bound with plain straps of iron, this relic exhibits the date 1589, the year following 
the dispersion of the Spanish Armada, when generosity towards the Protestant poor 
was harshly modified by the fines and penalties inflicted upon recusants and Papists. 

Among alms-boxes of subsequent dates to the above, the following may be 
noted: Bramford, Suffolk, 1591 ; Aylestone, Leicester, 1613; Clapham, Bedford, 1627; 
Sedbergh, Yorks., 1633; Manton, Rutland, 1637; Bletchley, Bucks., 1637; Lostwithiel, 
Cornwall, 1645; while the number of others remaining bearing various dates durmg 
the seventeenth century indicates that the custom of recording the years in which 
these receptacles of benevolence were made was a very general one. 

A little oak cupboard in the historical gate-house at Rye Plouse is worthy of 
mention, as it is dated 1659, and yet possesses the inlay of mother-o'-pearl and bone 
which is usually associated with the elaborate Dutch features which are seen on English- 
made pieces after the Restoration. The same advanced method of decoration may 
be noticed on an even earlier piece, dated 1653, in the Victoria and Albert Museum. 

Among the many curious confirmations of the tulipomania which spread from 
Holland to England in the seventeenth century, an interesting example may be 
cited in a portion of a chest-front in the little museum at Aylesbury. Towards the 
end of the seventeenth century the mama was probably at its height, and the long- 
forgotten owner of this piece, who was evidently something of an enthusiast, has 
recorded his predilection by adorning the box with conventional forms of tulips, 
accompanied by the legend, " 16 E H 92." 

Chairs do not appear to have been so commonly inscribed with the dates of 
their construction as coffers and chests, though a goodly number remain which have 
initials or other evidences of ownership attached to them, which in some cases may 
assist to determine the approximate date at which they were made. 

In the north transept of Oundle Church, Northants., is a curiosity in the shape 
of a chair of rude semi-Gothic form, bearing the inscription: — 

SUMPTVoAP HMATO IpApN LONDINENSIVMOA.D. <> 1576. 

This relic is so weird and unusual that to place any reliance on its style as 
appertaining to the period in which it was dated would be misleading. It should 
be compared with the more accepted example, dated 1 560, which stands in the chancel 
of Epworth Parish Church, Lincolnshire. 

The custom of attaching the year of their manufacture to articles of furniture 
was not confined solely to the woodwork of such pieces. Chairs with stuffed backs 
and seats were sometimes provided with such dates executed in needlework. A 
low-backed chair of this type in the Victoria and Albert Museum is embroidered 
with a floral pattern in coloured wools after the manner of a carpet, the date 1649 
appearing on its back. It is, of course, quite possible that the embroidery, which 
is worked on a canvas covering stretched over the seat and back of the chair, may 
have been carried out a few years after the framework, but the twisted rails of the 
latter seem to indicate a confirmation of the accompanying date, as spiral members 



A HISTORY OF OAK FURNITURE 39 

did not come really into vogue until about the middle of the century, though they 
may have been made in inconsiderable numbers at a slightly eeirlier date. 

During the Elizabethan and Jacobean eras, bedsteads were probably more 
frequently carved with the dates of their origin than any other article of furniture. 
The great increase in the area of glass for lighting purposes, which was an architectural 
feature of the former epoch, resulted in discomfort in the shape of draughts, rendering 
it necessary to construct the bedsteads of the upper and wealthy classes in such form 
that they could be, so to speak, utilised as a room within a room, the framework 
of the bedstead being further assisted by hangings. The luxury of the age thus 
resulted in the making of a great amount of splendid structures, which were so much 
prized that they were frequently left as heirlooms or specially mentioned in wills 
and bequests. There is little wonder that their possessors should desire to place 
the acquirement of such precious articles on record. Oak bedsteads bearing 
seventeenth-century dates not infrequently find their way into present-day sale- 
rooms, and generally command high prices, though appreciation of them is not 
invariably shown. In April, 1908, a fine specimen, carved with caryatids, foliage, 
and strapwork, bearing the date of 1634, was knocked down for no more than 
fifty-five guineas at Messrs. Christie, Manson and Woods'. Several specimens of 
dated bedsteads and cradles remain in the National Collection at South Kensington, 
but it is to be feared that some of them have not altogether escaped the hands of 
the restorer." The cradle illustrated in Plate LXV., an excellent specimen of such 
pieces in use among middle-class families, bears the following interesting inscrip- 
tion : — 

14th DAI October C.M.B. 1641. 

Though this cradle is in fine condition, the rockers are much worn, as if with 
constant use. 

Besides boxes, chairs, cupboards, and bedsteads, implements and household 
utensils were frequently carved with the years in which they were constructed. In 
1904 a very interesting discovery was made among some lumber in the Cambridge 
Corn Exchange, where an ancient standard bushel measure was unearthed, which, 
on examination, was found to be inscribed with the following words: — 

ELIZABETH DEI GRACIA ANGLIC FRANCIS ET HIBERNI^ 

REGINA, 
accompanied by the date 1601. Porringers and wooden platters inscribed with dates 
are also at times discovered, though rarely, in remote districts. 

Examples of furniture occasionally crop up on which the dates, originally 
genuine, have been altered or falsified in order to give a spurious value to such 
pieces. The " Kroger " chest in the Victoria and Albert Museum, on which the date 
has been transformed from 1803 to 1603, is a case in point, and has been dealt with 

* One bedstead, with finely moulded pillar-caps and inlaid panels at the back, is dated 1593. 



40 A HISTORY OF OAK FURNITURE 

in the author's work, Old Oak Furniture; but another piece, also at South Kensington, 
is not so well known, and seems to have escaped remark. The latter is an arm-chair 
of ash, from Vierlande, in Germany (No. 1113 — 1904), with turned spindle supports to 
the back and arms, and bearing on the back panel a carved rose and the incised 
inscription : — 

SOPHIA WULFFS 
ANNO 1795. 

The date, on examination, will be found to be in part an alteration, probably 
from 1825, in a presumable endeavour to delude on the part of its one-time owners. 

In conclusion to this cnapter, let it be said that dates have been, and still are, 
added and affixed to furniture and fittings of old buildings, not only of the 
approximate periods to which such examples belong, but also in most appallingly 
loose and incorrect ways. The characteristic Jacobean screen at Apethorpe Hall, 
Northants., exhibits on one of its panels the inscription: — 

VIRTVS SVPER OMNIA VINCAT 1364, 

being only antedated by some 250 or 300 years. 

The belfry ladder at Play den Church, Sussex, also shows an eleventh-century 
date, carved many years ago, but in Arabic characters. The spurious attribute on 
the Great Bed of Ware {temp. Elizabeth) has been dealt with in my book on Old 
Oak Furniture, and is of a character very different from the genuine date of the 
same epoch, with its cross-legged fours, which figures above the entry to the churchyard 
at Bray. 




Plate VI i 








c 

2: 



y. -^ -^ 







CHAPTER IX. 

Personal Experiences. 

I HAT this rhapter could easily be extended to equal length with the rest 
oi the book I have not the slightest doubt, and that it would not even 
then be the most wearisome portion I have also some views upon. But 
to be egotistical is a heinous fault to be accused of, and I will curtail 
these personal experiences — for this time at least. A good many adventures have 
been forgotten, but I shall hardly forget my first attempt to acquire a Gothic meuble. 
A youth affected with a singular passion for the pointed styles, I walked one April 
day, many years ago, into the Rue Eau de Robec in Rouen, in those days a veritable 
scrap depository of crazy odds and ends of furniture. The little bridges across 
the dirty ditch led straight up to dark, cavernous doorways in the old houses on 
the other side, contrasting sharply with the bright sunlight without. Inside one of 
these doorwa}s, and half -obscured in shadow, stood a tall press, or armoire, in which 
a number of flamboyant panels were discernible. I crossed the bridge and endeavoured 
to examine the cupboard, which was half-covered with a litter of various descriptions. 
Yes! the panels were of fifteenth-century flamboyant design, bearing the arms of 
France and Brittany on shields inserted among the tracery. Very much worn and 
rubbed they appeared, a great deal more so than the framing, which suggested 
replacement of a later date, but the whole piece was covered with a coat of dark- 
brown paint, obscuring the outer surface. Inside everything was conformable: the 
backs of the panels presenting a semblance of decay, while the stiles and transoms 
showed up more sharply cut and of newer surface. The proprietor made his 
appearance, and tilings began to look interesting. 

Monsieur was regarding the meuble Gothique. It was tres belle -piece, and 
very rare. The panels had been found in the Robec — which might account for their 
condition — the framing had been restored, but the piece was so very exceptional. 
The price was somewhere about five or six pounds, and I inwardly resolved that 
even if the framework was modern, the panels were well worth the money. But 
somehow the thing lingered. Were there any more such pieces? The proprietor 
would go and see. Left alone for a few minutes, something prompted me to scratch 
the face of one of the panels with my knife, and the bald truth was at once revealed. 
The beautiful flamboyant carvings were " squeezes " only, backed with old oak and 
painted on their fronts. 

I did not acquire that armoire, though within a few hours I was lucky enough 

41 



42 A HISTORY OF OAK FURNITURE 

to get hold of some really genuine panels of similar character, and at a price which 
the present-day collector might search for in vain. 

If you meet a hunting-man who asserts that he has never been thrown, don't 
believe him; similarly, if you are told by a collector that he has never once been 
taken in — well, you can say or think pretty much what you like about that person, 
whoever he may happen to be. The accounts of many subsequent failures to 
discriminate shall be passed over here. 

Belief is pretty general amongst amateurs that " finds " in the raw and untouched 
state are verging on extinction, so swept is the countryside for business purposes 
nowadays; but for those who have eyes to see and ears to hear, some very attractive 
discoveries yet remain to be gathered in by the true collector. Quite recently, while 
exploring tliat part of the Thames haven which Mr. Arthur Morrison, the novelist, 
has made so alluring and so peculiarly his own, I chanced across a humble and rather 
lonely cottage, which displayed a modest announcement as to the supplying of tea 
and hot water. On the little grass-plot in front of the dwelling were gathered 
together some three or four tables and a few chairs, intended for the use of such 
festive parties as happened to find their way to this quiet and secluded spot. The 
tables were of different shapes and sizes, mainly of the commonest order, all painted 
a blatant white, and the tops alike covered with American cloth ; but there was no 
mistaking the Carolean outline of one of them, with its pillar legs and moulded 
rails. The latter table, which had evidently been out in the weather for many years, 
had lost the best part of its feet through damp-rot, and had been elevated on splints 
of firewood in a very quaint manner; but it was of oak, and possessed a very elegant 
outline. An approach in due form was necessary. The owner of the establishment 
was engaged in thinning the hedges on his plot, and was summoned by his wife, 
who refused to do anything without the sanction of her better half. My enquiry- 
elicited the information that the male vendor of refreshments would cheerfully sell 
the table for 2s. in order to replace it with a new one. I paid an inclusive sum to 
cover transport, and was asked doubtfully if I cared for such old stuff. An 
affirmative answer produced an old pistol from the man's pockets. " I've seen this 
kickin' about in the hedge for weeks, but it's no use to me. You can have it for 
sixpence." So that little silver-mounted pocket " barker," engraved with the 
monogram " J. J.," had evidently lain concealed among the hedge-roots ever since 
some desperate affray or smuggling raid early in the nineteenth century. What lost 
histories are those of the Carolean table and hidden Georgian pistol, and what fresh 
imaginative romances could be weaved around them ! 

In 191 0, while sojourning at a town in Northern France, I detected some 
interesting looking antiquities in an establishment connected with carriage hiring. 
The articles were obviously a vendrc, but the assistant could not give me any 
information except that the proprietor was out at the time, and that he had gone to 
view a Gothic armoire, which was said to have come from the crypt of the old 
church. This sounded interesting. An appointment was arranged for the afternoon, 



A HISTORY OF OAK FURNITURE 43 

and my second visit unearthed M. le Proprietaire, one of the handsomest and finest 
men it has ever been my luck to set eyes on. I purchased a few things of no great 
account, and then made enquiries as to the mysterious nieiible Gotlnqite. It 
appeared that the individual who owned the piece was uncertain as to his habits, 
but my magnificent Frenchman would take me to the place where it remained. A 
short walk brought us to a neat old-fashioned house in a sunny square. No answer 
was returned to repeated applications at the door, but a neighbour who appeared 
volunteered the information that though the resident had been called away on 
business, she could take us inside. The hall and front room were partly filled up 
with two or three heavy, uninteresting armoires or presses, dating possibly from the 
reign of Louis XIV., and not remarkably desirable from the collector's point of 
view. My conductor interrogated the woman who had let us into the house, and then 
made his way to a room at the back. I have no notion as to whether he expected 
me to accompany him, but I went, and to my surprise entered a very complete 
"faker's" workshop. Amidst the tools and pickling stuffs were the dismantled 
sections of another ponderous Louis cupboard, which was in process of transition 
into a piece Gothiqiie. It was all very interesting. Such proceedings are a pity, 
but there was so much superfluous material in this case to work upon that no great 
amount of harm was done — certainly not to me ! 

It is not many collectors who have the good fortune to bring home veritable 
specimens of Gothic art after a day's casual outing in the country, but it has fallen 
to my lot several times, and always in the most unexpected manner. One Saturday, 
a few years ago, I visited a little town in the Eastern Counties, where communication 
with London was difficult and unsatisfactory. Trains were but few, and the 
connections were irritating, not to say thoroughly bad. I had tea at a cosy old- 
fashioned inn, after inspecting everything that was interesting in the place, and was 
preparing to return when I ran across a small general shop which had hitherto 
escaped my notice. In the window a very indifferent chest of " Charley " t)pe was 
exhibited among a scanty display of curios, but what interested me most was the 
sight of an undoubted fifteenth-century joint-stool, which was standing in a lop-sided 
fashion among some lumber down the adjacent yard. The good woman who answered 
my summons was very morose; she could not tell me the price of anything, as her 
husband had gone out, and they had disagreed about some sales she had effected. 
She refused to name a price for the stool or anything else until her husband came 
back, which I suspected would be from the nearest public-house. When he would 
return she did not know, and she would do absolutely nothing without the man. 
There was no help for it, waiting was my only chance. I went all over the town 
again, and finally returned to the inn for some supper. The last train for London 
left between nine and ten, and this would land me home at something past midnight, 
but I was determined to have that stool if possible. The little private room at the 
inn was illuminated by a very insufficient oil-lamp, and I sat and ate cold meat and 
read a weird book of occult stories till time was nearly up. Arriving once more at 



44 A HISTORY OF OAK FURNITURE 

the shop, it was at once apparent, from a wordy warfare going on, that the owner 
had returned in a heated and argumentative condition. He had not intended to 
sell the stool, but I could have it for live shillings, and he would carry it to the 
railway station for me if I would give him the price of a drink. Payment was duly 
made and accepted, and we started, accompanied by some flowers of speech from 
the woman as a send-off. On the way my man, who took matters in what seemed to 
me a very leisurely way, displayed a remarkable disposition to rest and tell me all 
about his family quarrels. The whistle sounded as we arrived outside the railway 
station. There was no time to get a ticket; my precious stool was bundled into a 
ramshackle carriage as the train was moving, and I fell in after it, the only passenger, 
tired out but successful. 




Plate IX. 



TYPES OF CONSTRUCTION 




¥//m//^///^/////////////S, 



Qi 



•*=*'=*l 



lfB^W?>A 





Type I. — the thirteenth-century coffer, front and ends. 




Type II. — the folrteenth-century coffer, with simulated pa.nelling. 



wmiiMm \ yi»)»»))H i, \ y)>iw»mA — \l'»)i»»»».i 








D 
D 
D 
D 
D 
c 



Type HI. — the panelled chest, with heavy fra.ming. 



Type IV. — the coffer with dovetailed edges. 



Plate X. 



■t^m 



JL'//////// 



'//////^//w/mm/z/^M^ 



w/;//////////MMmmmi> 





■^=^=Nl 



Type V. — klizabethan chest. 



\:\ . : : ■! ^..'t : , ' ". 7-~-\ 



YW^' 



Type VI. — Jacobean chest. 




Type VII. — the slab-ended box. 



1 





construction of cothic stool. 



Plate XI. 



PERIOD FURNITURE 




\N-««\T\^-e_ 



LATE TWELFTH-CENTURY COFFER, IN HINDRINGHAM CHURCH, NORFOLK. 




THIRTEENTH-CENTURV COFFER, JN BOSHAM CHURCH, SUSSEX 



Plate XII. 




PANEL FDOM A THIRTEENTH-CENTURY COFFRET. IN SAFFRON VVA1.DEN MUSEUM. 



^/ 



id 








FOOT OF THE BOSHAM COFFER. 



^ 




















LATE FOURTEEXTM-CEXTUKV COFFEK, IN YORK CATHEDRAL. 










.1 



m^ 












COFFER OF THE PE^PEN DICLLAR PERIOD. FORMEKI.Y 1 .\ SEDLESCO.MBK CHLRCH, SUSSEX. 
F;<0.\I A SKETCH CY RKV. E. GODDAKD. 1S35 (T. II. HORSFIELD'S -'COLMY OF SUSSEX'). 





LATE FOURTEENTH OR EARLY FIFTEENTH-CENTURY COFFER, IN CHICHESTER CATHEDRAL. 
LENGTH, S FT. 8 IN. ; WIDTH, 16 IN. ; DEPTH, 16 IN. 



Plate XIV. 




FOURTEE.VTH-CENTURY COFFRET (NOW IN POSSESSION 
OF THE MARQUIS OF GRANBV); AND A CHEST, WITH 
PARCHEMIN PANELS, EARLY SIXTEENTH CENTURY. 



vatT 7n.^UBnKL N>U 




FOURTEENTH-CENTURY BUTTRESSED COFFER, IN 
ST. JOHN'S HOSPITAL, CANTERBURY. 




THE CRADLE OF HENRY V.. NOW IN THE LONDON MUSEU.VI. 



Plate XV. 




FOURTEEXTH-CENTL-RY COFFEK, l.\ BRAXCEPETH CHURCH, XORTH LMBERLAXD. 
FROM A DRAWIXG BY MR. W. G. FOOTITT. 




LATE FOLKTEEXTH-CEXTLKY BOX. WESTPHALl AX. IX THE FOSSESSIOX OF MR. F. {.OKDOX ROE. 
THE STRAPS AXD IROXWORK GEXERALLY ARE TYPICAL OF THE PERIOD. THE TERMIXATIOXS OF 

THE HAXOLE ARE FORMED AS BOARS' HEADS. 



Plate XVI. 




FIFTEENTH-CENTURY PANEL. A RELIC OF THE CHOIR-FITTINGS OF 
YORK CATHEDRAL. SALVED FROM THE FIRE OCCASIONED BY THE 

INCENDIARY .MARTIN IN 1829. AUTHOR'S COLLECTION, 



Plate XVII. 



w 






I 



=s 



^-'^ ^^ 



>l< 



^•^ 




ITB 



I 



i 



tlii! 



EARLY SIXTEENTH-CENTURY SIDE-TABLE, VK^TOHIA AND ALBERT MUSEUM. 







^ 



SIXTEENTH-CENTURY DRAW-TABLE AND BENCH. FROM BROADUAV, NEAR ILMINSTER, SO.MERSET. 
VICTORIA AND ALBERT MUSLU.M. 



Plate XVIII. 



2^ 



■-i.^s(tfi^ - i.s-; i.' i 




FIFTEENTH-CENTURY CHEST. FRENCH. FROM THE COLLECTION OF THE LATE BARON SWINFEN OF CHERTSEV. 




L\TE FIFTEEN1 H-CENTUUY CUPHOARD. VIC'JODIA AND ALBERT MfSECM. 



Plate XIX. 




EARLY SIXTEKNTH-CENTURY DOLE-CUHBOARD. SAID TO HAVE COME FROM IVY CHURCH, AN OLD HOUSE AT 
ALDERBURV, NEAR SALISBURY. FORMERLY A MONASTERY, NOW IN RUINS. VICTORIA-AND ALBERT MUSEU.W. 




CHEST. FIRST HALF OF SIXTEENTH CENTURY. WFSTFH ALI AN. 

THE TRAY OF THIS CHESl' HAS A SMALL SECUET PLACE IN A DOUBLE BOTTOM. 



Plate XX. 




LATE FIFTEENTH OR EARLY SIXTEENTH-CENTURY CABlNhl. 

FOUND IN A BARN OF AN OLD MANOR-HOUSE NEAR BRADFIELD, SIIEI-FIELD. 




DON OF WOOD, COVERED WITH " CUIR-BOUI LLI, " BOUND WITH IRON STl^AFS. 
PRESENIED TO LINCOLN'S INN I\ 1^49 BY ITS THEN TREASURER, HENRY HEVDUN. 



Plate XXI. 





EXAMPLES OF TYKOLESE lUR.NlTUKE OF COTIilC DESIGN. AND A FRENCH OR FLEMISH CHEST 
CARVED WITH FLELRS-DE-LYS. 



Plate XXII. 





Plate XXIII. 




LATE OOTHIC COFFER. TE\1P. HEXRV Vll. OR HEXRV VIII. 

THE " HERRING-BO.NE " PATTERN IS TVPICAL OF THIS PERIOD; THE ROUNDED TOPS TO THE TRACERY ARE 

ALSO INDICATIVE; BOTH FEATURES WERE TO BE FOUND IN WOODWORK AT THE DAWN OF THE RE.NAISSANCE. 




UTE GOTHIC COFFER, IN THAXTEI) CHURCH, ESSE.\. 

THE METHOD OF CONSTRUCTION WITH SLAB ENDS. THE LATE OGEE a'rCHES. AND THE STRING-COURSES 

WITHOUT TERMINATIONS. ARE CHARACTERISTIC. 



Plate XXIV. 




CHEST. TE.MP. HENRY VIII. IN TM li POSSESSION OF COL. H. C. T. LITTLEDALE. 

THE C.\KVING ON THE PANELS IS CO.MPAUABLE WITH THAT AT ABIXOTON ABBEY, NORTHANTS. SEE PLATE XXXIV. 





TYPES OF TUDOU FLOUEHING FOUND ON FUKNITLRE OF THE PERIOD. 



Plate XXV. 




U 7-. 



ss 



■■a Z 

h f- i 

S g G 

■i o u 

. ... J 



.., S '■' 

K O X 

:: 2= H 




r -J a a 



5 y '-' 



z !« X -J 

C = Z H 

« o < 

!i: ~ H ^ 

> z ;S 

< I = J 

- -J; < C 

O ? M 2 



UD X - 

.. "■ s 

= X s 

■< H 25 

E r O 

O ? i 



Plate XXVI. 




COTHIC JOINT STOOI,, IN HREDK CHUKCH, SLSSi;\. 




FIFTEENTH-CENTURY COFFRET. 

FROM THE LATE WALTER WIT HALL'S COLLECTION. 



Plate XXVI 1. 




JOINT STOOL, IN SAFFRON WALDEN MLSEU.M. 




FRONT PANEL IN. THE ANCIENT CORPORATION COFFER FORMERLY KEPT IN THE MOOT HALL, IPSWICH. AND 
NOW IN CHRISTCHUKCH MANSION, IPSWICH. MEASUREMENTS OF PANEL, Igi IN. BY JjJ IN. 



Plate XXVIIl. 




GAMING TABLE, I 530, IN THE POSSESSION OF LORD DE LISLE. 

THIS IS. PHESUMABLY, THE SPECIMEN' ILLUSTRATED IN SHAWS "SPECIMENS OF 

ANCIENT EURNITUDE," WHICH DESCRIBES IT AS BEING THEN AT HILL HALL, ESSEX. 





GOTHIC JOINT STOOL. IN IMISSESSION OF 

THE HAROUESS OF GRANBY. 



JOINT STOOL. FROM A FIFTEENTH-CENTURY " BOOK 

OF HOURS," BODLEIAN LIBRARY (mS. CANON LIT.. 99). 



Plate XXIX. 




"GRACE BEFORE MEAT, " BY JAN STEEN. AT BELVOIR CASTLE.^ 

THIS INTERESTING PICTLRE SHOWS A LINEN-PANELLED CUPBOARD-SEAT, OF A DATE ANTERIOR TO THE PAIN'TING. 

IT IS A VERY FAITHFUL REPRESENTATION OF THE USE OF SUCH BEAUTIFUL PIECES. 



Plate XXX. 




MIDDLE OF THE SIXTEENTH-CENTURY CHEST FRONT. THE ROUNDELS, HEADS, AND ARABESQUES 

INDICATE STRONGLY THE INl-LUKNCE OF THE ITALIAN RENAISSANCE UPON FLEMISH CARVING OF THE PERIOD. 




LINEN-PANELLED CHEST. WITH LIGHT FRAMING. FRENCH OR FLEMISH. 

THE THINNESS OF THE STILES INDICATES A LATE PERIOD OF GOTHIC ART. 



EARLY SIXTEENTH CENTURY. 



Plate XXXI. 




FIFTEENTH-CEXTUKY PANELS. FRENCH, FROM ROUEN. SIXTEENTH-CENTURY P.4NEL. RHONE VALLEY OR THE SAVOY. 

IN THE AUTHOR'S COLLECTION. 



■■;■=-■■■<? 



Plate XXXII. 




.MID-SIXTEEXTH-CENTURY PANEL. NORTHERN FRENCH OR FLEMISH. AUTHOR'S COLLECTION. 



Plate XXXIIl. 




FRENCH CREDENCE OF RENAISSANCE DESIGN. IN THE COLLECTION OF MR. ERNEST WVTHES. 



Plate XXXIV. 



1 os.f=-^--i^^ 





















'I ,:,J" 





.E 



PKTOU THOMAS SIl.KSTED'S COFFEK. UATED I5IQ, IN SHANKLI.N CHUKCH, ISI.K OF WKIHT. 
FNOM SllAW'S "SPECIMENS CF ANCIENT FURNITURE." 




Vre^\l^\^^"=- 



JONGLEURS CARVED ON THE FRIEZE OF A ROOM, AT ABINCTON ABBEY, NORTHANTS. 
TEMF. HENRY VIII. 



Plate XXXV. 




THIS BEALTIHUL SPECIMEN OF A TUDOR COURT CUPBOARD CAME FROM OXFORD. IX KENT. 
WHERE IT WAS DISCOVERED IN AN OUT-HOUSE DOINO DUTY AS A RECEPTACLE FOR CHEESES. 
COLLECTION OF SIR EDWARD BARRY, BART. 



Plate XXXVI. 












n 



U ; 




c,; 



J 



FANKL. TEMP. HENKV Vlll. 
FROM THE WALTHAM ABBEY 
ROOM. VICTORIA AND ALBERT 
MUSEUM. 




THE ■•jester" PA.NEL, AT RYE HOUSE. TEMP. HENRY V!1I. 

ONE OF THE FEW AUTHENTIC REPRESENTATIONS IN WOOD OF 
A JESTER; ANOTHER BEING O.N SOME FRENCH PANELI.1N(",. A 
CAST OF WHICH IS AT SOUTH KENSINGTON. 
IN EACH CASE HUMOROUS EXPRESSIONS ABE INDICATED. 




DOOR. WALNUT. SOUTHERN 
FRENCH. 1526. VICTORIA 

AND ALBERT .MUSEUM. 



Plate XXXVII. 




CHEST FRONT. FRENCH. TEMP. FRANCOIS l' '' IN THE AUTHOR S COLLECTION. 




CUPBOARD, ILLUSTRATING THE DECADENCE OF THE LINEN PANEL. 
IN THE POSSESSION OF MR. T. C. PARKER. 



Plate XXXVIII. 




SIXTEENTH-CENTURY CASSONE. WALNUT. ITALIAN. 




'^'-""LAiv 



^j m ^ m ^ mm ^ m ^^ m ^: 



sl\teenth-ce;ntury cassone. walnut. itallw. j-rom the soulaces collection, 
victoria am) albert museum. 



Plate XXXIX. 





SIXTEENTH-CENTURY STATE CHAIR. 
VICTORIA AND ALBERT MUSEUM. 



STATE CHAIR. FRE.NCH (LYONS). 

FROM THE PIEKPONT MORGAN COLLECTION. 

FOR.MERLV IN THE ROUGIER COLLECTION. 



Plate XL. 




SIXTEENTH-CENTURY BUFFET, AT FONTAINEBLEAU. FRENCH. 



Plate XLI. 




IRON-BOLXD COFFER. PROBABLY LATE SIXTEENTH CEN'TURY. DISCOVERED IN A SECRET HIDING-PLACE 

AT INT.ATESTONE HALL, ESSEX. AND ALLLDED TO BY .MISS BRADDOX I.\ "LADY ACDLEY'S SECRET." 




SIXTEENTH-CENTURY IRON-BOUND COFFER. IN FRA.MLINGHAM CHURCH, SUFFOLK. 



Plate XLIl. 




SIXTEENTH-CENTLRY IRON-BOUND COFFER. 
AT ST. SWITHIN'S CHURCH. WORCESTER. 




IKON-BOUND COFFER. REPLACING ONE BROKEN INTO IN 1 59O. 

AT WEST GATE, WINCHESTER. 



Plate XLIU. 



=.^^. 




ALMS-BOX, DATED I5S9. IN DOVERCOLKT CHURCH, ESSEX. 



PLATE XLIV. 




SECRETAIRE, DATED 1594. THE UPPER PORTION IS OF CAMPHOR OR CYPRESS- 

WOOD, AND IS OF ITALIAN ORIGIN. THESE SECRETAIRES WERE OFTEN PROVIDED 

WITH A SUBSTRUCTURE OF OAK UPON THEIR IMPORTATION INTO ENGLAND, AS IN THE 
SPECIMEN SHOWN. AUTHOR'S COLLECTION. 



Plate XLV. 




•J ON M r z 
:r ... - o 

: - 3 « S^ 




mi 



ws^ 






^' 






fcs, 





Plate XLV] . 




s a 







; .v't.i ■n*.:' 



Plate XLVII'. 




Plate XLVIII. 




ELIZABETHAN 1)R AW-TA lil.K ; A VEKY 
ANt) BOG-OAK ON ITS STRETCHKKS. 



UKAUTII-UL SPECIMEN IN I-INE CONDITION, 
VICTORIA AND ALBERT ML'SELM. 



KICHLY INLAID WITH HOLLY 




EIGHT-LEGGED TABLES OF THE SIXTEENTH CENTURY ARE EXCEEDINGLY SCARCE. 



Plate XLIX. 




DRAW-TABLE OF USUSLAL FORM. FIRST QUARTER OF SEVENTEENTH CENTURY. 




ELIZABETHAN DK.WV-TAHLE. 




SEVENTEENTH-CENTURY DRAW-TABLE. DUTCH. IN POSSESSION OF .MR. ARTHUR L. RADFORD. 



Plate L. 




T. - 



2 S5 



c 



o 



Plate LI. 




ELIZABETHAN" PANELLING OF THE " NELSON KOOM, STAN HOTEL. <;KE-.T VARMOLTil. 




L^ 



NAME INSCRIBED ON A SEVENTEENTH-CENTL'RV TABLE IN THE FOSSESSIO.V OF 
MR. RICHARD H. MORTE.V, THE SAVOY, DENHAM, BUCKS. THE INSTONES WERE 
A YEOMAN FAMILY RESIDING I.NTHE NEIGHBOURHOOD DURING THE SEVENTEENTH 
CENTURY", THEIR DESCENDANTS DISPOSING OF THE TABLE circa lS6^. 



Plate Lll. 





^XvV)\';^ 



Plate LIII. 




BEDSTEAD-HEAD. YORKSHIRE. FIRST HALF OF SEVENTEENTH CE.NTLRV. 




ELIZABETHAN CHEST. I.\ THE POSSESSION OF .MR. L. W.iLFOKD, OF BLDLEIGH SALTERTO.N. 



Plate LIV. 




SIXTKENTII-CKNTUKV TABLE, SUTTON PLACE, SLKREV, 




SEVENTEENTH-CENTURY THAVELLINC, TABLE. IN THE AUTHOR S COLLECTION. 



Plate LV. 




SEVENTEENTH CENIURV FIREPLACE. IN CATHERINE OF 
ARAGON'S house, SHREWSBURY. 



J L 




.MECHANISM OF MOVABLE PILLAR HIDING KEYHOLE OF 
DUTCH CABINET. ' THE SHAFT OF THE PILLAR IS EBONISED 
SEE PLATE LVI. 



Plate LVI. 



DUTCH CflBINET , // C£NruflY 




SEVENTEENTH-CENTL-KV CABINET. DUTCH. IN THE .4UTH0K'S COLLECTION. 



Plate LVII. 




TWO CHAIRS. TEMP. CHARLES I. AND ELIZABETH. THE EARLIER CAME FRO.M RICHMOXD, YORKS. 

AND BEARS THE INITIALS OF WILLIA.M BLAND. WHO FLOURISHED IN THE REIGNS OF .MARY AND ELIZABETH. 
author's COLLECTION. 




DUG-OUT COFFER. IN BISHOP'S CLEEVE CHURCH. GLOUCESTER. 

the locks suggest FOURTEENTH OR FIFTEENTH-CENTURY WORK, WHICH IS 

THE ONLY INDICATION OF AGE. 



Plate LVIll. 





TWO INTERESTING BIBLE-BOXES, IN MR. TOWNROE'S COLLECTION. THAT CARVED WITH THE GLILLOCHE 

AND ROSE PATTERN IS JACOBEAN. THE OTHER IS RARER. SINCE IT BEARS THE CO.MMONWEALTH DATE 

OF 1632. THE LITTLE MASK UNDER THE LOCK-PLATE IS INTENDED TO REPRESENT CHARLES 1., SHOWING 

THE ORIGINAL OWNER OF THE BOX TO HAVE HAD ROYALIST SYMPATHIES. SUCH MASKS ARE SELDOM 

FOUND ON BIBLE-BOXES; THEY OCCUR USUALLY ON THE SO-CALLED " VORKSHl RE " TALL-BACKED CHAIRS, AND 
O.N THE HILTS OF ".MORTUARY" SWORDS. 



Plate LIX. 




CHAIR, DATED 1649. VICTORIA AND ALBERT MUSELM. 




BIBLE-BOX. FIRST HALF OF SEVENTEEKTH CENTL'RV. FROM COLCHESTER. 

IN THE POSSESSIO.V OF MR. F. GORDON ROE. 



Plate LX. 




_-*■• 












SEVENTEENTH-CENTURY CHEST. HORDEHS Ol- OXON ANU WARWICK. 

POSSIBLY OF SCOTTISH ORIGIN, 




LATE SIXTEENTH OR EARLY SEVENTEENTH-CENTURY CHEST. 



Plate l.Xl. 




FLEMISH CLPBOARD. SEVENTEENTH CENTLRV. 

THIS IS A TVPICAL SPECIMEN OF THE DUTCH RENAISSANCE ; MANY SIMILAR SPECIMENS. 

HAVE BEEN ATTRIBUTED ERRONEOUSLY TO OUR ELIZABETHAN CRAFTSMEN. 

VICTORIA AND ALBERT .MUSEUM. 



I'LAIE LXII. 




SPOON-CLPHOAKI) AND KMFK-UOX. JUNCTION OF THK SKVENTKENTH AND KU'.HTKENTH CENTL'KIES. 
AN INTERESTING SPECIMEN EOK PCRELV DOMESTIC LSE. AUTHOR'S COLLECTION. 



Plate LXIU. 




LATE SIXTEENTH OR EARLY SEVENTEENTH-C EXTLRV CHEST. FORMERLY L\ WINTER BOLR.NE HOLSE. NEAR BRISTOL. 




FIRST H-\LF OF THE SEVENTEENTH CENTURY. WEST SOMERSET. 




J.YCOBEAN CHEST, FROM CHLRCH STRETTO.N. SHROPSHIRE. 



Plate LXIV. 




SEVEXTEEXTH-CENTLRV MAROUETERIE CHEST. FROM MADAME BLANCHE MARCHESI S COLLECTION. 




RED CEDAR-WOOD COFFER, WITH PAINTED LID. ITALIAN. EARLY SIXTEENTH CENTURY. 



Plate LXV. 




SEifLE. SE\E.MEENTH CEXTLKV. DISCOVERED IN' A PL BLIC-HOLSE AT RIPPONDEN, 

0\ THE BORDERS OF YORKS. AND LANCS. 




CRADLE, DATED 164I. VICTORIA AND ALBERT MUSEUM. 



Plate LXVI. 




COFl-KK. IN FINGRIXCHOE CHLUCH, KSSKX. THE DATE (1684) IS PROBAHl.V AN AI)[)IT10N. 








flfTT* ^ 



COl-l-KK, MADE ]1V JAMES i;KM I'l-l S IN I'J^C). VICTOinA ANII Al.llEKT MLSEUM. 



Plate LXVII. 




LATK SIXTKENTH OR EARLY SEVENTEEXTH-CENTL'RV SLAB-END COFFER, WITH INCISED DECORATION. ELM. 

THE BORDLRE OF TRAILED CONVENTIONAL ROSE-LEAVES IS UNCO.M-MOX, AND RE.MINISCENT OF EARLIER TIMES. 




\J 



IVKllLLSi; CHEST, DATED 1624. PINEWOOD. IN THE COLLECTION OF MAJOR-GENERAL SIR 

GROVE, K.C.B. THIS TYPICAL SPECIMEN EXHIBITS SURVIVAL OF GOTHIC INFLUENCE. 



COLERIDGE 



Plate LXVIII. 



T^e\ 




iC^ 





V^^M 




€> 




* . ^ < 



CHEST OF DRAWERS. SECOXO HALF OF SEVENTEENTH CENTURY. EASTERN COUNTIES. 

author's COLLECTION. 



Plate LXIX. 




\,^\Y\:-. 



SPICE CUPBOARD. SECOND HALF OF THFl SEVE.NTEEXTH CEXTLKV. AUTHOR'S COLLECTION. 




CAHINF.T. INLAID WITH MOTH ER-O -PEARL AND IVORY. 
SECO-ND HALF OF THE SEVENTEENTH CENTURY. 



Plate LXX. 




SMALL SQUARE JACOBEAN STOOL. FROM 1>0L\D"S lilillX^E. KENT. 

author's collection. 









inscription carved on lid of oak coffer, formerly COiNTAINIXC 
DOCUMENTS OF THE FREE SCHOOL, AiND NOW IN THE COURT HOUSE, RYE. 



Plate LXXI. 





TRIANGLLAR CHAIR, FORMERLY AT CHESHUNT 
GREAT HOUSE. ATTRIBUTED TO CARDINAL 

WOLSEV. ACTUALLY SEVENTEENTH CENTURY. 



CAROLEAN STOOL. KNOLE. 





ARMCHAIR, INSCRIBED WAV.. 169Q. 
I-IIRMERLV IN COTK HOUSE. RRISTOL. 



EARLY SEVENTEENTH-CENTURY CHAIR 
FOUND IN CHELSEA. 



Plate LXXII. 





CHAIR WITH FIEKCKD BACK. FLEMISH. 

SEVENTEENTH CENTURY. FROM THE LATE 

WALTER WITHALL's COLLECTION. 



CHAIR. TEMP. CHARLES II. FROM THE 

LATE SIR EDWARD HOLDEN'S COLLECTION. 



Plate LXXIII. 




DAY-BED. TEMP. CHARLES II. FROM THE LATE SIR EDWARD HOLDENS COLLECTION'. 



Platic LXXIV. 



TWO CHAIRS FROM THE LATE SIR EDWARD HOLDEN'S COLLECTION. 





TEMP. WILLIAM AND MARY. LATE SEVENTEENTH CENTURY. 

THE BELL SHAPED MKMIIERS ON THE LEGS OF THE ARMCHAIR ARE TYPICAL OF THE TIME. 



Platk LXXV. 




STOOL, AT MIDHLRST PARISH CHURCH, IN'SCRIBED 

I — B 

1 6 89 

T — P 

ANU SHOWING A SURVIVAL OF THK GOTHIC FORM. 




LOCK-PLATE OX THE ING ATESTONE COFFER. SEE PLATE .\L1. 

FRO.M BUCKLER'S "TWENT\-T\VO CHURCHES OF ESSEX," 1856. 



Plate LXXVI. 




DUTCH BOX. SEVENTEENTH CENTURY. IN THE POSSESSION OF .MR. J. B. HARRIS-BURLAND. 




CARVED WINDOW-SILL IN .MAIDEN LANE, STAMFORD. SAVED FROM DESTRUCTION BY MR. H. F. TRAVLEN, F.R. I. B. A. 
THE ARMORIAL BEARINGS ARE THOSE OF THE DIGBYS. THIS SPECIMEN IS INCLUDED AS THE CARVING IS SO 

EXCEEDINGLY TYPICAL OF THE FIFTEENTH-CENTURY CRAFTSMAN'S ART. 




10^1- 



..V NK 



:Hi; ONTARIO C0LLEG5.i?F's- ^^ 

N ' Ro3 



i